A mobile manipulator human shared control method and system

Abstract

The application discloses a kind of mobile mechanical arm man-machine shared control method and system, including host computer remote control system and lower computer control system;Host computer remote control system includes main industrial computer, and the mechanical arm master hand of electric connection with industrial computer and shared controller;Lower computer control system includes embedded industrial computer, mechanical arm slave hand;The application realizes the dynamic change of shared control weight, so that fuzzy control joins the consideration of operation environment restriction problem, reduces the complete dependence on algorithm by shared control, so that mechanical arm can still complete corresponding task in the case where operation environment is limited.

Description

Technical Field

[0001] This invention relates to a robot control method, and more particularly to a human-machine shared control method and system for a mobile robotic arm. Background Technology

[0002] With the continuous development of robot control technology, the level of robot intelligence is also constantly improving. When facing more complex working environments or tasks involving delicate operations, traditional control methods, including remote control and autonomous control, cannot meet the corresponding task requirements. Therefore, human-robot shared control is introduced into robot systems, which combines direct operator control commands with partial autonomous robot control to enhance the robot's adaptability in different environments. Current robot shared control systems generally consist of direct operator control, partial autonomous robot control, and a combination of both. Direct operator control mostly uses manual control commands, where the operator directly controls the robot via keyboard, remote sensing, or other devices, but this can sometimes be affected by environmental factors and system latency. Autonomous robot control employs path planning, navigation, and localization technologies, heavily relying on corresponding algorithms. The shared control portion mostly adopts multi-fuzzy rule-based robot control systems, with fuzzy control and neural networks being typical methods.

[0003] For shared control systems of robots (robotic arms), existing shared control methods suffer from limitations in automation levels and sensor technology. When environmental constraints exist, fuzzy control schemes used to control the robotic arm result in low task completion rates. For example, when the robot's working environment changes, the adaptive performance of the shared control system is generally limited by the limitations of intelligent fuzzy control algorithms. Summary of the Invention

[0004] Purpose of the invention: To address the above problems, this invention proposes a human-machine shared control method and system for a mobile robotic arm. It realizes the dynamic change of shared control weights, incorporates the consideration of operating environment constraints into fuzzy control, and reduces the complete dependence on algorithms through shared control, so that the robotic arm can still complete the corresponding tasks even when the operating environment is limited.

[0005] Technical solution: The technical solution adopted in this invention is a human-machine shared control system for a mobile robotic arm, including a host computer remote operating system and a slave computer control system;

[0006] The host computer remote control system includes a main industrial computer, a robotic arm master hand electrically connected to the industrial computer, and a shared controller; the slave computer control system includes an embedded industrial computer and a robotic arm slave hand.

[0007] The robotic arm is equipped with sensors and a 3D camera. The sensors collect first state information and feed it back to the joint change output module in the embedded industrial control computer for processing. The joint change output module outputs the joint change amount of the robotic arm. The first state information includes the degree of freedom state of the end effector of the robotic arm. The 3D camera is used to collect image information.

[0008] The embedded industrial control computer transmits the joint changes and image information to the main industrial control computer. The main industrial control computer processes the image information to obtain the distance between the robotic arm's hand and the target object. The fuzzy controller in the main industrial control computer outputs shared control coefficients based on the second state information and fuzzy control rules, and sends them to the shared controller. The second state information includes the joint changes of the robotic arm's hand and the distance between the robotic arm's hand and the target object.

[0009] The industrial control computer sends master-slave control commands to the shared controller; the embedded industrial control computer sends autonomous control commands to the shared controller; the shared controller outputs shared control commands and sends them to the robotic arm slave, controlling the robotic arm slave to move;

[0010] The industrial control computer is equipped with a human-machine interface.

[0011] The system also includes a mobile platform for carrying a robotic arm from the hand and moving the robotic arm to the corresponding position.

[0012] The shared controller outputs a shared control command, the shared control command S c for:

[0013] S c (η)=(I o -I A )η+I A

[0014] Among them I o Master-slave control commands, I A η represents the robot's autonomous control command, and η is the shared control coefficient.

[0015] The master and slave hands of the robotic arm are configured in a master-slave isomorphic manner. For the master-slave isomorphic robotic arm, a joint space mapping control method is adopted, which synchronously maps the joint angles of each joint of the master controller to the slave joint space in a certain proportion to control the operation of the slave hand of the robotic arm.

[0016] The robotic arm adopts a point-to-point trajectory planning method from the autonomous control of the hand, and uses a PID controller for trajectory tracking; the point-to-point trajectory planning is carried out using a fifth-order polynomial.

[0017] A shared control method applied to the aforementioned human-machine shared control system for a mobile robotic arm, wherein the joint change output module calculates the joint change amount based on the degrees of freedom of the end effector of the robotic arm's hand, using the following formula:

[0018]

[0019] In the formula, ΔX is the change matrix of the end effector, ΔQ is the joint change matrix, q is the channel quantity, including the rotation angle and displacement distance of the joint; n is the number of joints, m is the number of degrees of freedom of the end effector; f is the mapping between each element in the degree of freedom vector group of the end effector and the channel quantity q, denoted as x=f(q);

[0020] The fuzzy rules of the fuzzy controller include: the larger the joint change, the larger the shared control coefficient; the smaller the joint change, the smaller the shared control coefficient.

[0021] The controller uses the Mamdani fuzzy inference method to implement fuzzy inference.

[0022] Its fuzzy implication relationship The fuzzy set of the joint changes Q of the robotic arm and the distance L between the robotic arm's end effector and the target object. and The formula for calculating the fuzzy implication relation of the Cartesian set is as follows:

[0023]

[0024]

[0025] in and Let Q represent the quantized membership degrees of the robotic arm joint changes and L between the robotic arm end effector and the target object, respectively, in their respective universes of discourse for each fuzzy set. This represents the degree of membership of the implication relations of the input object when performing fuzzy inference. To determine the membership degree of the fuzzy set on the universe of discourse after the shared control coefficients are quantized, R i For each fuzzy control rule, there is an implication relation.

[0026] The total output of fuzzy logic reasoning is:

[0027]

[0028] In the formula, where The intersection of the membership degrees of each fuzzy set after quantization of the input object, where i is the number of fuzzy rules.

[0029] For the membership function of the fuzzy set of the joint change Q of the robotic arm and the distance L between the robotic arm and the target object, a combination of trapezoidal membership function and triangular membership function is adopted. When the fuzzy set is in the middle degree, the triangular membership function is adopted, while the trapezoidal membership function is adopted at the two extreme degrees.

[0030] For the membership function of the fuzzy set with shared control coefficient η, a combination of triangular membership function and S-shaped membership function is adopted. The triangular membership function is used when the fuzzy set is in the middle degree, while the S-shaped membership function is used at the two extreme degrees.

[0031] Beneficial Effects: Compared with existing technologies, this invention has the following advantages: This invention adopts a control system with a master-slave architecture and establishes a human-machine shared control method, namely, human-machine shared control based on operator master-slave control and robotic arm trajectory tracking. Specifically, for the human-machine shared control strategy, a combination of a joint change output device and a fuzzy controller is used, enabling the robotic arm to dynamically adjust control weights according to the surrounding environment as it moves towards the target object. Furthermore, the proposed shared control method fully considers the unique characteristics of the robotic arm compared to other robots, namely, the joint changes that occur during movement, including angle and displacement changes. By adjusting the amount of joint changes, the robotic arm control system can optimize the robotic arm's operating posture to complete the task in environments with limited operating space, relying on operator master-slave control, without increasing the complexity of the system algorithm. Attached Figure Description

[0032] Figure 1 This is the overall framework of the human-machine shared control system for mobile robotic arms;

[0033] Figure 2 This is a diagram showing the relationship between shared control coefficients and operating modes;

[0034] Figure 3 This is a block diagram of the human-machine shared control system for a mobile robotic arm;

[0035] Figure 4 This is a block diagram of a master-slave control system;

[0036] Figure 5 This is a block diagram of PID control.

[0037] Figure 6 This is a diagram of the fuzzy controller structure;

[0038] Figure 7 This is a flowchart of the human-machine shared control process for a mobile robotic arm. Detailed Implementation

[0039] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0040] The mobile robotic arm human-machine shared control system described in this invention adopts a master-slave structure. The master computer remote operating system consists of a computer and a master robotic arm (master hand). The industrial control computer is responsible for the human-machine interface, and the operator sends master-slave control commands to the controller through the master hand. In the slave computer control system, an embedded industrial control unit forms an intelligent control system responsible for controlling the robotic arm (slave hand) located on the mobile platform according to the shared control commands. Simultaneously, it feeds back environmental information collected by sensors to the master computer, displays it to the operator through the human-machine interface, and sends autonomous control commands. The overall framework of the mobile robotic arm human-machine shared control system is as follows: Figure 1 As shown. The working principle of this shared control system is as follows: after the upper and lower computer communicate via a wired connection, the operator views the robot arm's posture and issues master-slave control commands through a human-machine interface. Simultaneously, the intelligent control system outputs autonomous control commands based on environmental information and the robot arm's own state. Finally, the shared controller, after receiving the master-slave control commands, autonomous control commands, and environmental information, outputs shared control commands. The specific shared control process is as follows: Figure 7 As shown. The shared control command S of the output design is... c for:

[0041] S c (η)=(I o -I A )η+I A (1)

[0042] Among them I o Master-slave control commands, I A Here, η represents the robot's autonomous control command, and η is the shared control coefficient. From equation (1), we can see that the magnitude of the shared control coefficient η determines the weight of the operator's and the robot arm's autonomous control in shared control. The magnitude of the η coefficient determines the control mode, which can be categorized into autonomous control, operator-master-slave control, and human-machine shared control, such as... Figure 2 As shown.

[0043] (1) When η = 0, the operation mode is the autonomous control mode of the robotic arm. The operation command output by the shared controller is the autonomous control of the robotic arm. At this time, the operator does not participate in the control of the robotic arm. The robotic arm performs autonomous control based on the environmental information of visual feedback.

[0044] (2) When 0 < η < 1, the operation mode is the human-machine shared control mode, that is, the operation command output by the shared controller is a weighted fusion command of the operator's master-slave control command and the robot's autonomous control command. In this mode, when η = 0.5, the operator and the robot each have half of the control authority. As the value of η increases, the control weight of the operator increases.

[0045] (3) When η = 1, the operation mode is the operator master-slave control mode. The operation instructions output by the shared controller are the operator master-slave control instructions. At this time, the robotic arm is completely controlled by the operator, that is, the operator controls the robotic arm (slave hand) through the master hand.

[0046] Based on the aforementioned mobile robotic arm human-machine shared control system, this invention adopts a human-machine shared control method that combines operator control based on master-slave control and robotic arm autonomous control based on trajectory tracking. The main content is to design an effective human-machine shared control strategy that integrates the operator's master-slave control commands with the robotic arm's autonomous control commands, and realizes dynamic weight allocation during the control process.

[0047] The principle is to first integrate operator state, robotic arm state, and environmental information to establish a fuzzy controller model for receiving data, and then output a shared control coefficient η according to the designed shared control rules. Then, based on the shared control coefficient η, the operator's master-slave control command I0 and the robotic arm's autonomous control command I1 are executed. A Perform fusion and output shared control command S c This achieves shared human-machine control of the robotic arm. The human-machine shared control block diagram for the mobile robotic arm is shown below. Figure 3 As shown.

[0048] According to the above-mentioned human-machine shared control block diagram of the mobile robotic arm, the human-machine shared control method of the robotic arm is specifically divided into three parts: (1) operator master-slave control; (2) autonomous control of robotic arm trajectory tracking; (3) human-machine shared control combining the two.

[0049] (1) As a preferred technical solution of the present invention, a master-slave control method is adopted to realize the operator's control of the robotic arm. The master-slave control system consists of the following parts: master control terminal, control system, slave control terminal, and feedback system. The specific master-slave control system is as follows: Figure 4 As shown. The operator transmits control signals from the master hand to the slave hand via the control system.

[0050] The master and slave arms adopt a master-slave isomorphic configuration, meaning their structures are identical, differing only in size. This isomorphic robotic arm employs a joint space mapping control method, synchronously mapping the angles of each joint of the master controller to the slave joint space at a specific ratio. Simultaneously, a 3D camera measures the distance between the robotic arm's end effector and the target object, feeding this information back to the operator via the control system. This allows the operator to control the slave arm in real-time during operation, optimizing control commands based on environmental changes.

[0051] (2) As a preferred technical solution of the present invention, the robotic arm grasping unit is mounted on the robot moving unit and can reach the corresponding position along with the robot moving platform. The robotic arm adopts a point-to-point trajectory planning method and cooperates with a PID controller for trajectory tracking, specifically as follows:

[0052] (21) The robotic arm trajectory planning adopts a point-to-point planning method. Point-to-point planning, as a method of joint space planning, is used when given constraints on the start and end points, and the goal is to find the trajectory connecting these two points. This technical solution uses a fifth-order polynomial for planning, and its general form is:

[0053] θ(t)=a0+a1(tt s )+a2(tt s ) 2 +a3(tt s ) 3 +a4(tt s ) 4 +a5(tt s ) 5 (2)

[0054] The six unknowns require six constraints, namely the angular displacement, angular velocity, and angular acceleration at the starting and ending points. The constraint equations are as follows:

[0055]

[0056] Solving the system of equations yields six unknowns, where T = (t e -t s ).

[0057]

[0058] Substituting the parameters into the fifth-degree polynomial, we obtain the desired trajectory.

[0059] (22) Following the trajectory planning of the robotic arm, this invention proposes using a PID controller for trajectory tracking. The PID controller consists of proportional control, integral control, and derivative control. It automatically adjusts the three parameters of the PID controller based on the system error and the rate of change of the error, thereby achieving better control performance. The specific structure is as follows: Figure 5 As shown.

[0060] Proportional control (P): The output of a proportional controller is proportional to the difference between the current state and the target state. A larger proportional ratio results in a faster approach to the target value, but it also increases the risk of overshoot. A smaller proportional ratio reduces overshoot, but the response time becomes much longer. Furthermore, when only proportional control is used, the system output exhibits a steady-state error.

[0061] Integral control (I): The output of the integral controller is proportional to the integral of the input error signal. Integral control is generally used to eliminate the steady-state error of a system.

[0062] Differential control (D): The output of differential control is proportional to the rate of change of the input error signal. Differential control can be used to reduce the overshoot of pure proportional control.

[0063] In addition, regarding PID controllers, the following formula exists:

[0064]

[0065] Where e(i), i = 0, 1, 2, 3, ..., k, k is the corresponding systematic error, k p k i k d These are the parameters for the PID controller. The three parameters are adjusted to a suitable set of values ​​based on the task. The adjustment process generally involves: first adjusting k... i k d Set k to zero and adjust the P controller separately until the system response reaches an optimal result, i.e., the response speed is acceptable and the overshoot is small. Then fix k. p Adjust the integral controller k d .

[0066] (3) As a preferred technical solution of the present invention, in order to realize human-machine shared control and reasonably allocate the control weight between the operator and the robotic arm, the present invention proposes to set up a method that combines the joint change output device with the fuzzy controller to finally obtain the shared control coefficient η.

[0067] 1) Joint change output device

[0068] The purpose of the joint change output device is to obtain the joint change by inversely transforming the changes in the end effector of the robotic arm through the inverse operation of the robotic arm's motion equation, and then output it to the fuzzy controller. The specific method is as follows:

[0069] The rotation angle and displacement distance of the joint are collectively referred to as channel quantities, denoted as q. The target state is denoted as a vector group X = (x1, x2, x3, ..., x...). m ) T Let X be the vector set of angular velocity and linear velocity of the end effector. Each element x can be obtained from q, denoted as x = f(q). By introducing the Jacobian matrix, ΔQ (joint variation) is related to ΔX (end effector). The joint degrees of freedom are obtained by deriving the following formula:

[0070]

[0071] Extending the above equation, taking the total differential of each term in X, and then taking the partial derivative of the right side of the equation with respect to all q, we get the following expanded equation:

[0072]

[0073] Where m is the number of terms in X, representing the number of degrees of freedom of the end effector, and the number of columns in the matrix equals the number of joints in Q, as simplified below:

[0074] ΔX m×1 =J m×n (q)·ΔQ n×1 (8)

[0075] From the above formula, we can see that in the inverse operation, the degree of freedom of each joint can be calculated by using the degree of freedom of the end effector, that is, by multiplying ΔX by the inverse (pseudo-inverse) of the Jacobian matrix, we can obtain the change of Q.

[0076] 2) Fuzzy controller

[0077] A fuzzy logic controller is employed, using the joint change Q and the distance L between the robotic arm's end effector and the target object as inputs. Fuzzy rules are established based on expert experience from practical work, and the two types of information are fused to finally output shared control coefficients. The design of the fuzzy logic controller in this invention is as follows:

[0078] (31) Determine the structure of the fuzzy controller

[0079] This invention employs a fuzzy controller structure as follows: Figure 6 As shown, its working principle is as follows: First, the precise input quantity is fuzzified and converted into a corresponding fuzzy quantity. Then, fuzzy inference is performed according to the principle of induction. Finally, the inference result is clarified to obtain the precise output value, which is the value of η.

[0080] (32) Determine the fuzzy union of input and output.

[0081] The basic universe of discourse for the change in the joint of the robotic arm is [0, 1]. Its fuzzy universe of discourse is set to {0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1}. The change in the joint of the robotic arm is divided into 5 fuzzy sets {VS, s, M, B, VB}, which correspond to {very small, small, medium, large, very large} respectively.

[0082] The basic universe of discourse for the distance L between the end effector of the robotic arm and the target object is [0, 1]. Its fuzzy universe of discourse is set to {0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1}, with a quantization factor of 0.1. The distance between the end effector of the robotic arm and the target object is divided into 5 fuzzy sets {VC, C, M, F, VF}, corresponding to {very close, close, medium, far, very far}, respectively.

[0083] The basic universe of discourse for the shared control coefficient η is [0, 1]. Its fuzzy universe of discourse is set to {0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1}. Based on the magnitude of the shared control coefficient η, it is divided into 5 fuzzy sets {VS, S, M, B, VB}, which correspond to {very small, small, medium, large, very large} respectively.

[0084] (33) Membership function of fuzzy set

[0085] For the joint changes Q of the robotic arm and the distance L between the robotic arm's end effector and the target object, a combination of trapezoidal and triangular membership functions is used. Trapezoidal membership functions are used at both ends, while triangular membership functions are used in the middle. For the shared control coefficient η, a combination of triangular and S-shaped membership functions is used. Taking the joint variable Q of the robotic arm as an example, when the fuzzy set is {S, M, B}, triangular membership functions are used; when the fuzzy set is {VS, VB}, trapezoidal membership functions are used. This forms the membership function for the joint change Q of the robotic arm. The membership function for the distance L between the robotic arm's end effector and the target object is similarly constructed.

[0086] For the membership function of the fuzzy set of shared control coefficient η, a combination of triangular membership function and S-shaped membership function is adopted. When the fuzzy set is {S, M, B}, triangular membership function is used, while S-shaped membership function is used at {VS, VB}, thus forming the membership function of shared control coefficient η.

[0087] (34) Establishing fuzzy rules

[0088] Fuzzy rules are designed based on the following principles:

[0089] 1. When the distance between the end effector of the robotic arm and the target object is far, the robotic arm is mounted on the mobile platform and is still in the process of approaching the target object. At this time, the operator controls the mobile platform.

[0090] 2. As the robotic arm gradually approaches the target object, due to the inherent limitations of the robotic arm structure and the autonomous path planning algorithm, the robotic arm may experience shaking and the target position may become unreachable. In this case, the shared control coefficient η should be increased to increase the operator's weight in the shared control, and the operator's decision-making ability at the upper level should be used to adjust the path of the robotic arm.

[0091] 3. When the distance between the end effector of the robotic arm and the target object is very close, in order to avoid the operator's tension affecting the control accuracy, the shared control coefficient η should be reduced and the proportion of autonomous control of the robotic arm in the shared control should be increased, and autonomous control should be completed by relying on the robotic arm trajectory tracking.

[0092] 4. When the joint change is large, considering the limited workspace of the robotic arm, the shared control coefficient η should be increased to increase the operator's proportion in the shared control. That is, the master-slave control should be used to limit the joint change to avoid collisions.

[0093] 5. When the joint change is small, the shared control coefficient η is reduced to give full play to the advantages of the robotic arm's autonomous control.

[0094] (35) Fuzzy Reasoning and Defuzzification

[0095] This invention employs the Mamdani method for fuzzy inference, and its fuzzy implication relation (Q, L) can be obtained through fuzzy sets. and The Cartesian set is obtained by taking the joint variation Q of the robotic arm and the distance L between the robotic arm's end effector and the target object as a fuzzy set. and The calculation is performed as shown in formula (6).

[0096]

[0097]

[0098] in and These represent the membership degrees of the various fuzzy sets on the universe of discourse of the two input objects after quantization. The membership degree of the fuzzy set on the universe of discourse of the output object after quantization.

[0099] The total output of fuzzy logic reasoning is shown in Formula 7:

[0100]

[0101] Substituting the result of formula 6 into formula 7, we obtain U. * Finally, U is analyzed using the average maximum membership method. * The process is refined to determine the shared control coefficient η.

[0102] Specifically, using the above-mentioned control system and fuzzy rules, the execution steps of shared control are as follows:

[0103] a. First, determine the position of the target object and the initial position of the robotic arm arbitrarily in the given environment;

[0104] b. Based on the trajectory planning algorithm, design the path for the robotic arm to reach the target object;

[0105] c. Based on the trajectory tracking algorithm, the robotic arm autonomously tracks the planned trajectory.

[0106] d. Determine the distance between the robotic arm and the target object, divide the distance, and divide the working area into the operator's hand control area, the robotic arm trajectory tracking area, and the non-graspable area;

[0107] e. In the operator's manual control area, the operator adopts a master-slave control method; in the robotic arm trajectory tracking area, the robotic arm relies on autonomous control; in the non-graspable area, the robotic arm does not receive control commands.

[0108] f. The shared controller receives master-slave control commands and autonomous control commands, weights and fuses the two commands to obtain the shared control command, and outputs it to the robotic arm;

[0109] g. The robotic arm receives shared control commands and eventually reaches the target object's location.

Claims

1. A human-machine sharing control system for a mobile robotic arm, characterized in that: This includes a host computer remote operating system and a slave computer control system; The host computer remote control system includes a main industrial control computer, a robotic arm master hand electrically connected to the main industrial control computer, and a shared controller; the slave computer control system includes an embedded industrial control computer and a robotic arm slave hand. The robotic arm is equipped with sensors and a 3D camera. The sensors collect first state information and feed it back to the joint change output module in the embedded industrial control computer for processing. The joint change output module outputs the joint change amount of the robotic arm. The first state information includes the degree of freedom state of the end effector of the robotic arm. The 3D camera is used to collect image information. The embedded industrial control computer transmits the joint changes and image information to the main industrial control computer. The main industrial control computer processes the image information to obtain the distance between the robotic arm's hand and the target object. The fuzzy controller in the main industrial control computer outputs shared control coefficients based on the second state information and fuzzy control rules, and sends them to the shared controller. The second state information includes the joint changes of the robotic arm's hand and the distance between the robotic arm's hand and the target object. The master industrial computer sends master-slave control commands to the shared controller; the embedded industrial computer sends autonomous control commands to the shared controller; the shared controller outputs shared control commands and sends them to the robotic arm slave hand to control the movement of the robotic arm slave hand. The main industrial control computer is equipped with a human-machine interface; The joint change output module calculates the joint change amount from the degrees of freedom of the end effector of the robotic arm, using the following formula: ； In the formula, This is the change matrix of the end effector. Let represent the joint variation matrix, where q is the channel quantity, including the joint rotation angle and displacement distance; n represents the number of joints, and m is the number of degrees of freedom of the end effector; f represents the mapping between each element in the end effector's degree-of-freedom vector group and the channel quantity q, denoted as . ; The fuzzy rules of the fuzzy controller include: the larger the joint change, the larger the shared control coefficient; the smaller the joint change, the smaller the shared control coefficient. The controller uses the Mamdani fuzzy inference method to implement fuzzy inference; The formula for calculating fuzzy implication relations is: ； ； in and Let Q represent the quantized membership degrees of the robotic arm joint changes and L between the robotic arm end effector and the target object, respectively, in their respective universes of discourse for each fuzzy set. This represents the degree of membership of the implication relation of the input object in fuzzy reasoning. To determine the membership degree of the fuzzy set on the universe of discourse after the shared control coefficients have been quantized. For each fuzzy control rule, there is an implication relation. The total output of fuzzy logic reasoning is: ； In the formula, This is the total output of fuzzy logic reasoning. The intersection of the membership degrees of each fuzzy set after quantization of the input object, where i is the number of fuzzy rules.

2. The mobile robotic arm human-machine sharing control system according to claim 1, characterized in that: The system also includes a mobile platform for carrying a robotic arm from the hand and moving the robotic arm to the corresponding position.

3. The mobile robotic arm human-machine sharing control system according to claim 1, characterized in that: The shared controller outputs a shared control command. for: ； in Master-slave control commands For robot autonomous control commands, For shared control coefficients.

4. The mobile robotic arm human-machine sharing control system according to claim 1, characterized in that: The master and slave hands of the robotic arm are configured in a master-slave isomorphic manner. For the master-slave isomorphic robotic arm, a joint space mapping control method is adopted, which synchronously maps the joint angles of each joint of the master controller to the slave joint space in a certain proportion to control the operation of the slave hand of the robotic arm.

5. The mobile robotic arm human-machine sharing control system according to claim 1, characterized in that: The robotic arm adopts a point-to-point trajectory planning method from the autonomous control of the hand, and uses a PID controller for trajectory tracking; the point-to-point trajectory planning is carried out using a fifth-order polynomial.

6. The mobile robotic arm human-machine sharing control system according to claim 1, characterized in that: For the membership function of the fuzzy set of the joint change Q of the robotic arm and the distance L between the robotic arm and the target object, a combination of trapezoidal membership function and triangular membership function is adopted. When the fuzzy set is in the middle degree, the triangular membership function is adopted, while the trapezoidal membership function is adopted at the two extreme degrees.

7. The mobile robotic arm human-machine sharing control system according to claim 1, characterized in that: For the membership function of the fuzzy set with shared control coefficient η, a combination of triangular membership function and S-shaped membership function is adopted. The triangular membership function is used when the fuzzy set is in the middle degree, while the S-shaped membership function is used at the two extreme degrees.

Images