Safety control method and system for master-slave robot arm teleoperation

Abstract

The application discloses a kind of safety control method and system of master-slave mechanical arm remote operation, belong to robot remote operation technical field.The method includes: the Sigmoid nonlinear correction of main arm PWM signal is carried out to improve input fidelity;Real-time acquisition slave arm joint angle and limit, calculate safety margin and dynamically generate the mapping gain factor that reduces with the reduction of safety margin;After the angle of corrected main arm is compressed by Sigmoid, the current angle of slave arm is fused to generate target instruction according to the gain factor weighting;After target instruction limiting, it is issued to slave arm driver.The application can actively prevent joint movement over-limit, avoid emergency stop trigger, while improving remote operation smoothness and trajectory accuracy.

Description

Technical Field

[0001] This invention relates to the field of robot teleoperation technology, and more specifically, to a real-time control method and system for ensuring the safety of slave arm movement and the smoothness of operation under the condition of heterogeneous hardware between master and slave robotic arms. Background Technology

[0002] In hazardous or remote scenarios such as nuclear emergency response and deep-sea operations, master-slave remote operating systems are often used. A typical system consists of a master-end interactive device (such as a servo-driven robotic arm) and a slave-end high-precision industrial robotic arm (such as the Franka EmikaPanda) connected by a communication network. The operator issues action commands from the master end, and the slave-end robotic arm replicates them in real time.

[0003] However, significant hardware heterogeneity often exists between the master and slave ends. The joint motion range (e.g., 0°~270°) of the low-cost servo-driven robotic arm on the master end, and the nonlinear characteristics such as dead zone and hysteresis in its PWM control signal, contradict the strict, asymmetric physical joint limits (e.g., -π / 6~+π / 6) and high safety requirements of the slave-end industrial robotic arm. Using traditional static linear mapping methods will lead to serious problems:

[0004] Illegal instruction generation: When the master arm instruction approaches its limit, the linear mapping is prone to generating a target angle that exceeds the physical limit of the slave arm, causing the slave arm to trigger an emergency stop, report an error, or even suffer hardware damage.

[0005] Lack of proactive safety mechanisms: The mapping process is an open loop, making it impossible to perceive the real-time status of the slave arm. Even when the slave arm is close to the danger zone, it still transmits all commands, resulting in a high cognitive load for the operator and making it easy for the task to be interrupted due to negligence.

[0006] Poor operational fidelity: The nonlinear characteristics of the servo motor are not compensated, and the slight vibrations or jumps of the main arm are directly mapped to the abrupt movements of the slave arm, reducing the smoothness of operation and the success rate of the mission.

[0007] Existing research largely focuses on kinematic consistency optimization, failing to embed "joint-level physical safety" as a core constraint into the real-time control loop. Therefore, there is an urgent need for a safety control method that can dynamically sense the slave arm's state, adaptively adjust the mapping relationship, and prevent over-limit commands from the source. Summary of the Invention

[0008] (a) Purpose of the invention

[0009] This invention aims to solve the problems of joint over-limit, trajectory distortion and lack of active safety protection caused by static linear mapping in existing master-slave heterogeneous teleoperation. It provides a safety control method and system based on nonlinear dynamic remapping, enabling operators to safely, intuitively and with high fidelity remote control of high-precision slave arms through a low-cost master arm.

[0010] (II) Technical Solution

[0011] To achieve the above objectives, the present invention adopts the following technical solution:

[0012] In a first aspect, the present invention provides a safety control method for master-slave robotic arm teleoperation, applied to a teleoperation system including a master-end interactive device, a slave-end industrial robotic arm, a communication network, and a controller. The method includes the following steps:

[0013] Step S1: Collect the original control signals of each joint of the main terminal interactive device, perform nonlinear correction on the original control signals, and obtain the corrected main arm joint angles.

[0014] Step S2: Real-time acquisition of the current angle and physical limit of each joint of the slave industrial robotic arm;

[0015] Step S3: Based on the current angle and the physical limit, calculate the safety margin of each joint, and dynamically generate the mapping gain factor of the joint according to the safety margin, wherein the mapping gain factor decreases as the safety margin decreases.

[0016] Step S4: After compressing the corrected main arm joint angle using the Sigmoid function, the angle is weighted and fused with the current angle of the slave industrial robot arm according to the corresponding mapping gain factor to generate the target joint angle command.

[0017] Step S5: Limit the target joint angle command to ensure that it is within the physical limit of the corresponding joint, and send the processed command to the driver of the slave industrial robot arm.

[0018] Secondly, the present invention provides a teleoperation system for implementing the above-described method, comprising a master-end interactive device, a slave-end industrial robotic arm, a communication network, and a controller. The controller includes at least one processing unit and a memory, the memory storing a computer program. When the computer program is executed by the processing unit, it implements the various steps of the safety control method.

[0019] (III) Beneficial Effects

[0020] Compared with the prior art, the beneficial effects of the present invention are:

[0021] 1. By performing nonlinear correction on the main arm PWM signal including Sigmoid function processing, the dead zone and hysteresis characteristics of the servo motor at both ends of the stroke are effectively compensated, so that the input angle of the main arm more accurately reflects the operator's intention, improving command fidelity and operation smoothness.

[0022] 2. By calculating the safety margin of the slave arm joint in real time and dynamically generating a mapping gain factor that decreases as the safety margin decreases, the influence of the master arm command is automatically weakened when the slave arm approaches the physical limit. This avoids generating over-limit commands at the control source, eliminates the risk of emergency stop triggering and mechanical overload, and forms an active safety protection mechanism.

[0023] 3. By weighting and fusing the corrected and compressed master arm angle with the current slave arm position according to dynamic gain, and combining this with a final hard limit, a safety command generation strategy with "soft boundary" characteristics is formed. This allows the slave arm to smoothly decelerate and tend to stop when approaching the limit region, rather than abruptly or with a hard collision, significantly improving motion continuity and the naturalness of human-machine interaction.

[0024] 4. The method of the present invention does not rely on expensive force feedback equipment or complex kinematic calibration. It can achieve safe and accurate remote operation in a heterogeneous system composed of a low-cost master arm and a high-requirement industrial slave arm through intelligent dynamic remapping at the software level, effectively ensuring hardware safety and improving task completion efficiency. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein:

[0026] Figure 1 This is a schematic diagram illustrating the composition of a teleoperation system provided in an embodiment of the present invention.

[0027] Figure 2 This is a hardware connection diagram provided for an embodiment of the present invention.

[0028] Figure 3 The flowchart for angle mapping and command generation provided in the embodiments of the present invention is shown.

[0029] Figure 4 This is a schematic diagram comparing the traditional static linear mapping with the remote limit dynamic remapping strategy of this invention.

[0030] Figure 5 This is a schematic diagram comparing the traditional static linear mapping with the near-limit dynamic remapping strategy of this invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] The terms "first," "second," etc., used in this specification are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, the first object can be one or more. Furthermore, in the specification, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0033] This invention provides a safe control method for master-slave robotic arm teleoperation, applicable to a teleoperation system including a master-end interactive device, a slave industrial robotic arm, a communication network, and a controller. The method includes the following steps:

[0034] Step S1: Collect the original control signals of each joint of the main terminal interactive device, perform nonlinear correction on the original control signals, and obtain the corrected main arm joint angles.

[0035] Step S2: Real-time acquisition of the current angle and physical limit of each joint of the slave industrial robotic arm;

[0036] Step S3: Based on the current angle and the physical limit, calculate the safety margin of each joint, and dynamically generate the mapping gain factor of the joint according to the safety margin, wherein the mapping gain factor decreases as the safety margin decreases.

[0037] Step S4: After compressing the corrected main arm joint angle using the Sigmoid function, the angle is weighted and fused with the current angle of the slave industrial robot arm according to the corresponding mapping gain factor to generate the target joint angle command.

[0038] Step S5: Limit the target joint angle command to ensure that it is within the physical limit of the corresponding joint, and send the processed command to the driver of the slave industrial robot arm.

[0039] In step S1, the master-end interaction device is a servo-driven robotic arm controlled by a pulse width modulation signal.

[0040] The nonlinear correction includes:

[0041] The original control signal is linearly normalized to the [0,1] interval;

[0042] The normalized signal is compressed using the Sigmoid function.

[0043] The compressed signal is mapped to a preset physical angle range of the master terminal interaction device to obtain the corrected main arm joint angle.

[0044] In step S2, the slave industrial robotic arm is a high-precision servo-driven industrial robot with asymmetric physical limits.

[0045] In step S3, the first joints in Safety margin at all times Calculated using the following formula:

[0046] ;

[0047] in, The first end of the industrial robotic arm joints in The current perspective at any given moment; , For the first Physical limitations of joints.

[0048] In step S3, the first joints in Mapping gain factor at time Dynamically generated using the following formula:

[0049] ;

[0050] in, This is the preset reference gain coefficient; This is the preset decay time constant.

[0051] In step S4, the weighted fusion generates the target angle command. The formula is:

[0052] ;

[0053] in, This is the value obtained after compressing the corrected main arm joint angle using the Sigmoid function; This is the current perspective.

[0054] The present invention also provides a safety control system for master-slave robotic arm teleoperation, for implementing the aforementioned safety control method for master-slave robotic arm teleoperation, comprising:

[0055] The main interactive device is used to generate operation commands;

[0056] Slave-end industrial robotic arms are used to perform remote tasks;

[0057] A communication network connects the master-end interactive device and the slave-end industrial robotic arm for real-time data interaction;

[0058] The controller is connected to both the host interaction device and the communication network, and is configured to execute the security control method.

[0059] In another embodiment, the present invention also provides a safety control system for master-slave robotic arm teleoperation, comprising:

[0060] The master-side interactive device is used to generate joint control signals;

[0061] From the end-effector industrial robotic arm, each joint has physical limits; communication network; and

[0062] The controller is connected to the main interactive device and the communication network via signals respectively;

[0063] The controller includes at least one processing unit and a memory. The memory stores a computer program, which, when executed by the processing unit, controls the controller to perform the following steps:

[0064] The joint control signals of the main terminal interactive device are collected and nonlinearly corrected to obtain the corrected main arm joint angle.

[0065] Obtain the current joint angles and physical limits of the slave-end industrial robotic arm;

[0066] The safety margin is calculated based on the current angle and physical limit, and a mapping gain factor that decreases as the safety margin decreases is dynamically generated.

[0067] After the corrected main arm joint angle is compressed using the Sigmoid function, it is then weighted and fused with the current angle of the slave end according to the mapping gain factor to generate the target angle command.

[0068] The target angle command is sent to the slave industrial robotic arm after being subjected to amplitude limiting processing.

[0069] When the computer program is executed by the processing unit, it controls the controller to... Generate the mapping gain factor .

[0070] The following detailed description of the safety control method for remote operation of master-slave robotic arms provided by the present invention is based on specific embodiments.

[0071] This embodiment provides a safety control method for master-slave robotic arm teleoperation, applicable to, for example... Figure 1 The system shown. Combined with... Figure 2 As shown, the system includes a master-end interaction device (in this embodiment, a 7-DOF servo robotic arm EVO-UArmConfig3 with ZP10D servos for the joints), a slave industrial robotic arm (Franka Emika Panda robot), a gigabit Ethernet communication network, and an Intel NUC minicomputer (running Ubuntu 22.04 and ROS Noetic) as the controller. The controller communicates with the master and slave arms at a frequency of 1kHz via the ROS topic mechanism.

[0072] The control signal for the ZP10D servo is a PWM wave with a period of 20ms and an effective pulse width range of 500μs to 2500μs, corresponding to 0° and 270° of the output axis, respectively. Each joint of the Franka servo has asymmetric physical limits; for example, the limit of joint 2 is... This parameter has been preloaded into the ROS parameter server.

[0073] The control flow in this embodiment is as follows: Figure 3 As shown, the specific steps are as follows:

[0074] Step 1: Main arm signal acquisition and nonlinear correction.

[0075] The controller reads the raw PWM values ​​of 7 servo motors in real time via serial port. First, it is linearly normalized to the [0,1] interval and then hard-limited:

[0076] ;

[0077] ;

[0078] in, =500, =2500 corresponds to the minimum and maximum effective PWM values ​​of the servo motor, respectively; This is the linear normalized value of the original PWM signal, the purpose of which is to convert the hardware-level PWM signal into the concept of "relative position" to facilitate subsequent calculations; The first parameter is a normalized value after being hard-limited. The purpose is to force all possible values ​​to fall within the 0,1 range, preventing accidental mutations outside this range from affecting subsequent values. Then, the... A hard limit is applied to ensure it remains within the range [0.0, 1.0] to prevent abnormal inputs from causing overflows in subsequent calculations. This step converts the physical signal into a dimensionless variable and prevents abnormal inputs, providing a stable basis for subsequent processing.

[0079] Next, the normalized signal Sigmoid nonlinear compression is performed to accurately fit the dead zone and hysteresis characteristics of the servo motor at both ends of its stroke, resulting in compressed variables. :

[0080] ;

[0081] Among them, the shape adjustment coefficient Set the value to 10. This step significantly improves the resolution of the main arm input in the boundary region and effectively suppresses abrupt changes in commands caused by dead zone transitions, resulting in a smoother operating feel and a smoother response from the slave arm.

[0082] Finally, compress the variables. Mapping to the physical angle range of the main arm (0°~270°), the corrected main arm joint angles are obtained. :

[0083] = ;

[0084] This step ensures that the main arm commands have clear physical meaning and eliminates systematic deviations caused by PWM-angle nonlinearity.

[0085] Step 2: Real-time sensing of arm status.

[0086] The embedded controller obtains the first data in real time from the slave industrial robotic arm via a communication network. Current angle of the joint (Unit: radians) and its physical limits [ , ].

[0087] Step 3: Safety margin calculation and dynamic gain generation.

[0088] Calculate the first Joint safety margin This value represents the distance from the arm's current pose to the nearest physical boundary:

[0089] ;

[0090] This step enables real-time and accurate quantification of the safety status of the trailing arm movement, providing crucial input for subsequent safety decisions.

[0091] Based on the safety margin, the mapping gain factor is dynamically generated. :

[0092] ;

[0093] in, =0.8 is the reference gain. =0.3 is the decay time constant; this design makes it possible when When it approaches 0, It automatically approaches 0, thereby significantly reducing the impact of the master arm command on the slave arm, forming a "soft boundary" effect.

[0094] Step 4: Dynamic remapping and security instruction integration.

[0095] First, the corrected main boom angle Perform sigmoid function compression again to obtain This is to further smooth the instructions. Subsequently, weighted fusion is performed to generate the target angle:

[0096] ;

[0097] The purpose of the weighted fusion is to: [in areas far from the limiting region] When the value is approximately 0.8, the slave arm primarily follows the movement of the master arm; when approaching the limit area ( When →0), the arm gradually “locks” the current position to avoid being forcibly pulled into the illegal area, thus achieving a smooth transition rather than a hard collision.

[0098] See Figure 4 As shown, in traditional static linear mapping (as indicated by the dashed line in the figure), the master arm angle and the slave arm angle have a fixed proportional relationship. When the master arm command approaches its limit, the target angle of the slave arm generated by the mapping is very likely to exceed its physical limit, leading to an emergency stop. However, the dynamic remapping strategy adopted in this invention (as indicated by the solid line) reduces the mapping gain (i.e., the slope of the curve in the figure gradually decreases) when the slave arm approaches the limit, so that the target angle of the slave arm smoothly approaches the limit boundary, thereby avoiding the generation of over-limit commands and achieving "soft boundary" safety protection.

[0099] See also Figure 5 As shown, when the arm joint is close to its physical limit, traditional static linear mapping still transmits the full arm command (dashed line), which can easily lead to exceeding the limit. The method of this invention (solid line) significantly suppresses the influence of the arm command in the danger zone by calculating the safety margin in real time and dynamically adjusting the gain. The arm target angle is "locked" near the safe position, thereby achieving smooth deceleration and natural stopping, avoiding mechanical impact and emergency stop triggering.

[0100] Step 5: Command security limit and issuance.

[0101] right Implement hard limits to ensure it is strictly within [[ , The data is transmitted to the slave arm driver via the Franka Control Interface (FCI) at a frequency of 1 kHz. This step serves as a final safety measure, ensuring the validity of the output command even if the fusion result slightly deviates from the specified limits due to numerical errors.

[0102] In safe areas, the slave arm primarily follows the master arm's movements, ensuring operational sensitivity and intuitiveness. In dangerous critical areas, the slave arm's movements gradually "freeze" near its current position, providing a smooth transition rather than abrupt stops. This avoids hard collisions and sudden stops while providing the operator with clear force feedback (by reducing the perceived boundary through the slave arm's response).

[0103] Finally, the limited amplitude The command is encapsulated as a JointPosition message and sent to the Franka FCI module via ROS Topic ( / franka_ros_interface / joint_position_command). The FCI parses the command at a frequency of 1kHz and drives the servo motor. The entire control cycle (steps S1 to S5) is completed within 1 millisecond, forming a high-frequency, real-time safety closed loop. When the operator drags the UARM robotic arm at the master end, Franka smoothly follows; when the master arm attempts to push Franka joint 2 towards +0.6 rad (exceeding +0.5236 rad), the system... The speed will automatically drop below 0.1, allowing Franka to slowly approach the upper limit and then stop, thus avoiding triggering Franka's Joint Limit Violation.

[0104] It should be further explained that the algorithm parameters (e.g., k=10, The values ​​(=0.8, τ=0.3) were obtained through experimental calibration and stored in the ROS YAML configuration file, which supports online adjustment.

[0105] Experiments show that in a continuous one-hour teleoperated grasping task, the traditional linear mapping method triggers an average of 2.3 Franka joint over-limit emergency stops per hour, while the method of this invention achieves zero over-limit events. At the same time, the root mean square error of the end-point trajectory tracking is reduced from 8.7 mm to 5.1 mm, which significantly improves operational safety and task accuracy.

[0106] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0107] Furthermore, it should be noted that the scope of the methods and systems in the embodiments of the present invention is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. In addition, features described with reference to certain examples may be combined in other examples.

[0108] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the scope of protection of the present invention, and all of these forms are within the protection scope of the present invention.

Claims

1. A safety control method for master-slave robotic arm teleoperation, applied to a teleoperation system including a master-end interactive device, a slave-end industrial robotic arm, a communication network, and a controller, characterized in that, The method includes the following steps: Step S1: Collect the original control signals of each joint of the main terminal interactive device, perform nonlinear correction on the original control signals, and obtain the corrected main arm joint angles. Step S2: Real-time acquisition of the current angle and physical limit of each joint of the slave industrial robotic arm; Step S3: Based on the current angle and the physical limit, calculate the safety margin of each joint, and dynamically generate the mapping gain factor of the joint according to the safety margin, wherein the mapping gain factor decreases as the safety margin decreases. Step S4: After compressing the corrected main arm joint angle using the Sigmoid function, the angle is weighted and fused with the current angle of the slave industrial robot arm according to the corresponding mapping gain factor to generate the target joint angle command. Step S5: Limit the target joint angle command to ensure that it is within the physical limit of the corresponding joint, and send the processed command to the driver of the slave industrial robot arm. In step S3, the first joints in Safety margin at all times Calculated using the following formula: ； in, The first end of the industrial robotic arm joints in The current perspective at any given moment; , For the first Physical limitations of joints; In step S3, the first joints in Mapping gain factor at time Dynamically generated using the following formula: ； in, This is the preset reference gain coefficient; This is the preset decay time constant.

2. The safety control method for remote operation of a master-slave robotic arm according to claim 1, characterized in that, In step S1, the nonlinear correction includes: The original control signal is linearly normalized to the [0,1] interval; The normalized signal is compressed using the Sigmoid function. The compressed signal is mapped to a preset physical angle range of the master terminal interaction device to obtain the corrected main arm joint angle.

3. The safety control method for remote operation of a master-slave robotic arm according to claim 1, characterized in that, In step S4, the weighted fusion generates the target joint angle command. The formula is: ； in, This is the value obtained after compressing the corrected main arm joint angle using the Sigmoid function; From the current perspective; 4. The safety control method for remote operation of a master-slave robotic arm according to any one of claims 1 to 3, characterized in that, The master-end interaction device is a servo-driven robotic arm controlled by pulse width modulation signals.

5. The safety control method for remote operation of a master-slave robotic arm according to any one of claims 1 to 3, characterized in that, The slave-end industrial robotic arm is a high-precision servo-driven industrial robot with asymmetric physical limits.

6. A safety control system for master-slave robotic arm teleoperation, used to implement the safety control method for master-slave robotic arm teleoperation as described in any one of claims 1 to 5, characterized in that, include: The main interactive device is used to generate operation commands; Slave-end industrial robotic arms are used to perform remote tasks; A communication network connects the master-end interactive device and the slave-end industrial robotic arm for real-time data interaction; The controller is connected to both the host interactive device and the communication network.

7. A safety control system for master-slave robotic arm teleoperation, used to implement the safety control method for master-slave robotic arm teleoperation as described in claim 1, characterized in that, include: The master-side interactive device is used to generate joint control signals; From the end-effector industrial robotic arm, each joint has physical limits; communication network; as well as The controller is connected to the main interactive device and the communication network via signals respectively; The controller includes at least one processing unit and a memory. The memory stores a computer program, which, when executed by the processing unit, controls the controller to perform the following steps: The joint control signals of the main terminal interactive device are collected and nonlinearly corrected to obtain the corrected main arm joint angle. Obtain the current joint angles and physical limits of the slave-end industrial robotic arm; The safety margin is calculated based on the current angle and physical limit, and a mapping gain factor that decreases as the safety margin decreases is dynamically generated. After the corrected master arm joint angle is compressed using the Sigmoid function, it is then fused with the current slave angle according to the mapping gain factor to generate the target joint angle command. The target joint angle command is sent to the slave industrial robotic arm after being subjected to amplitude limiting processing.

8. The safety control system for remote operation of a master-slave robotic arm according to claim 7, characterized in that, When the computer program is executed by the processing unit, it controls the controller to... Generate the mapping gain factor .

Images