A dual-arm pose decoupling teleoperation system and method for single-master remote control of dual-slave arms
By using a single master hand to remotely control two slave arms, the posture decoupling teleoperation system solves the problems of high operational difficulty and excessive force in the coordinated movement of the two arms, achieving precise and flexible teleoperation and ensuring the safety of the target.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG UNIV
- Filing Date
- 2024-01-30
- Publication Date
- 2026-05-26
AI Technical Summary
Existing teleoperation technology is difficult to operate when both arms move in coordination, and it is easy to generate excessive force, which affects the safety and accuracy of the target.
A dual-arm posture decoupling teleoperation system with a single master hand remote control of dual slave arms was designed. The master hand controller controls the posture and joint angle of the left and right slave robotic arms respectively. Admittance control is used to achieve active compliance of the task target, and a dual-arm end spring damping model is used to prevent excessive force.
It achieves precise and flexible coordination of dual-arm remote operation, reduces operating costs, ensures the safety and smooth handling of the target object, and avoids damage.
Smart Images
Figure CN117754546B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of teleoperation technology, and in particular to a dual-arm pose decoupling teleoperation system and method for remotely controlling two slave arms with a single master arm. Background Technology
[0002] The statements in this section merely refer to the background art related to this invention and do not necessarily constitute prior art.
[0003] Teleoperation is a very old concept, referring to the control and operation of objects or systems at a distance via remote means. It is commonly used in scenarios requiring human intervention, where operators interact and control the target through remote devices or interfaces. For high-risk industries and work scenarios where direct on-site operation is unsuitable for humans, teleoperation has a very broad application prospect and scope, offering advantages such as high safety, remote operation capability, high precision and flexibility, and enhanced human capabilities.
[0004] Early research focused primarily on military and industrial applications, such as operations in nuclear and hazardous environments. In these fields, teleoperators can help humans avoid direct contact with dangerous environments, thus protecting operator safety. Over time, research on teleoperators has expanded to other fields, such as medical, space exploration, and rescue missions. In the medical field, teleoperated surgical systems have been developed, enabling doctors to perform precise surgeries remotely. In space exploration, teleoperators are used for spacecraft maintenance and assembly. In rescue missions, teleoperators can be used for search and rescue, rescuing trapped personnel, or handling hazardous materials.
[0005] However, existing teleoperation technologies mostly involve one master teleoperator controlling one slave robotic arm. This teleoperation method can meet the independent operation tasks of a single arm, but when it involves the coordinated movement of two arms, especially when a closed kinematic chain is formed between the two arms, the simultaneous remote control of two slave robotic arms by two master teleoperators poses a huge challenge to the operator's skill level. A slight mistake can generate huge forces on the rigid body that the two arms are acting on, which is not conducive to achieving the operation objectives of this invention. Summary of the Invention
[0006] To address the requirements for coordination, accuracy, and flexibility in teleoperation of dual-arm robots, this invention provides a dual-arm pose decoupling teleoperation system and method with a single master arm remotely controlling two slave arms.
[0007] On the one hand, a dual-arm pose decoupling teleoperation system with single master hand remote control of dual slave arms is provided, including:
[0008] Single master robotic arm and dual slave robotic arm; dual slave robotic arm includes: left slave robotic arm and right slave robotic arm;
[0009] The single master arm robotic arm is equipped with a master arm controller. The master arm controller performs decoupled teleoperation of the left or right slave arm, sending teleoperation commands to the left or right slave arm controller. The left or right slave arm controller controls the left or right arm to adjust its position according to the teleoperation commands, so that the left or right arm reaches the vicinity of the work target. The left or right slave arm controller adjusts the posture of the left or right arm according to the teleoperation commands, so that the left or right arm grasps the end of the work target.
[0010] The master controller performs collaborative pose decoupling teleoperation of the two slave robotic arms, sending teleoperation commands to the left and right slave robotic arm controllers. The left and right slave robotic arm controllers control the pose and joint angles of the left and right slave robotic arms respectively to adjust the position and attitude of the work target. The master controller achieves active compliance of the center position of the work target based on admittance control.
[0011] On the other hand, a dual-arm pose decoupling teleoperation method for single master-hand remote control of dual slave arms is provided, including:
[0012] The main controller receives instructions from the target of the handling operation;
[0013] The master controller performs a decoupled teleoperation of the left slave robot arm's posture, sending teleoperation commands to the left slave robot arm controller. The left slave robot arm controller controls the left arm to adjust its position according to the teleoperation commands, so that the left arm reaches the vicinity of the work target. The left slave robot arm controller adjusts the posture of the left arm according to the teleoperation commands, so that the left arm can grasp the left end of the work target.
[0014] The master controller performs a decoupled teleoperation of the right slave robot arm's posture, sending teleoperation commands to the right slave robot arm controller. The right slave robot arm controller controls the right arm to adjust its position according to the teleoperation commands, so that the right arm reaches the vicinity of the work target. The right slave robot arm controller adjusts the posture of the right arm according to the teleoperation commands, so that the right arm can grasp the right end of the work target.
[0015] The master controller performs collaborative pose decoupling teleoperation of the two slave robotic arms, sending teleoperation commands to the left slave robotic arm controller and the right slave robotic arm controller. The left slave robotic arm controller and the right slave robotic arm controller control the pose and joint angles of the left and right slave robotic arms respectively, so as to achieve the adjustment of the position and posture of the work target.
[0016] The master controller is based on admittance control to achieve active compliance in the center position of the work target.
[0017] The above technical solution has the following advantages or beneficial effects:
[0018] 1. The designed pose decoupling teleoperation algorithm can perform pose decoupling teleoperation on the end effector of a single robotic arm, controlling the position and attitude separately, making the entire teleoperation process more precise and flexible.
[0019] 2. The master end can use a single master hand to remotely control the coordinated operation of both arms, which greatly reduces the cost of operation. At the same time, the posture and position of the target can be controlled separately, making the remote operation of both arms more coordinated, flexible and convenient.
[0020] 3. The designed double-arm end spring damping model can ensure the smooth handling of the target and prevent excessive force from acting on the target, which could damage it. Attached Figure Description
[0021] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention.
[0022] Figure 1 This is a schematic diagram of the electrical connection relationship between the attitude decoupling teleoperation master arm and the two slave arms in this invention.
[0023] Figures 2(a) and 2(b) are schematic diagrams of the mechanical structure of the main hand and the mechanical structure of the end sphere in this invention.
[0024] Figure 3 This is a schematic diagram of the coordinated movement of the two arms in this invention.
[0025] Figure 4 This is the overall flowchart of the pose decoupling teleoperation method for single master hand remote control of dual slave arms in this invention.
[0026] Figure 5 This is a flowchart of the single-arm pose decoupling teleoperation algorithm in this invention.
[0027] Figure 6 This is a flowchart of the dual-arm cooperative pose decoupling teleoperation algorithm in this invention.
[0028] Among them, 1. First joint; 2. Second joint; 3. Third joint; 4. Fourth joint; 5. Fifth joint; 6. Sixth joint. Detailed Implementation
[0029] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0030] Example 1
[0031] As shown in Figures 2(a) and 2(b), this embodiment provides a dual-arm pose decoupling teleoperation system with a single master arm remotely controlling dual slave arms, including: a single master arm and dual slave arms; the dual slave arms include: a left slave arm and a right slave arm;
[0032] The single master arm robotic arm is equipped with a master arm controller. The master arm controller performs decoupled teleoperation of the left or right slave arm, sending teleoperation commands to the left or right slave arm controller. The left or right slave arm controller controls the left or right arm to adjust its position according to the teleoperation commands, so that the left or right arm reaches the vicinity of the work target. The left or right slave arm controller adjusts the posture of the left or right arm according to the teleoperation commands, so that the left or right arm grasps the end of the work target.
[0033] The master controller performs collaborative pose decoupling teleoperation of the two slave robotic arms, sending teleoperation commands to the left and right slave robotic arm controllers. The left and right slave robotic arm controllers control the pose and joint angles of the left and right slave robotic arms respectively to adjust the position and attitude of the work target. The master controller achieves active compliance of the center position of the work target based on admittance control.
[0034] Furthermore, the single master hand robotic arm includes: a first base, a first joint 1 mounted on the first base, a second joint 2 mounted on the first joint, the second joint 2 being connected to a third joint 3 via a first connecting rod, and the third joint 3 being connected to an end effector ball device via a second connecting rod;
[0035] The end-sphere device includes: a first columnar body, a fourth joint 4 mounted on the first columnar body, a sphere mounted on the fourth joint 4, a fifth joint 5 mounted inside the sphere, and the fifth joint 5 connected to a sixth joint 6 via a third connecting rod;
[0036] The left and right robotic arms have the same structure. The left robotic arm includes a second base connected in sequence. The second base is connected to the end effector through several sets of joint and link assemblies. Each set of joint and link assemblies includes joints and links connected in sequence.
[0037] Furthermore, such as Figure 1 As shown, the single master arm is equipped with a master arm controller, which is connected to the master arm encoder, motor and buttons respectively;
[0038] The master controller is connected to the left slave robot controller and the right slave robot controller via a CAN network.
[0039] The left-hand robotic arm controller is also connected to the left-hand encoder, the left-hand six-dimensional force sensor, and the left-hand voltage-current converter; the left-hand voltage-current converter is connected to the left-hand hydraulic cylinder through the left-hand electro-hydraulic servo valve; the left-hand hydraulic cylinder is connected to the left-hand robotic arm body; the left-hand encoder and the left-hand six-dimensional force sensor are connected to the left-hand robotic arm body.
[0040] The right-hand robotic arm controller is also connected to the right-hand encoder, the right-hand six-dimensional force sensor, and the right-hand voltage-current converter; the right-hand voltage-current converter is connected to the right-hand hydraulic cylinder through the right-hand electro-hydraulic servo valve; the right-hand hydraulic cylinder is connected to the right-hand robotic arm body; the right-hand encoder and the right-hand six-dimensional force sensor are connected to the right-hand robotic arm body.
[0041] As should be understood, Figures 2(a) and 2(b) show schematic diagrams of the mechanical structure of the master hand and the end effector ball used in this invention, which have a total of six rotary joints. The control system is placed in the base. The first joint is mounted on the base, the second joint is directly mounted on the components of the first joint, and the third joint is connected to the second joint and the end effector ball to the third joint by long connecting rods. The first three joints are used to determine the position of the end effector ball in space, and the joint information is obtained through a 32-bit high-precision encoder. The end effector ball is mainly composed of two short cylindrical bodies and a main ball, and the various parts are connected in series. The end effector ball integrates four degrees of freedom and six buttons. The built-in micro encoder can be used to read joint data. The fourth, fifth, and sixth joints can be used to control the posture of the slave arm, and the buttons can be used to switch the working mode of the posture decoupling master hand.
[0042] like Figure 1 The diagram illustrates the electrical connection between the master arm and the two slave arms in this invention. The master arm and the two slave arms are connected via a CAN network. The master arm and the two slave arms communicate through the CAN network to transmit information such as joint angles and end effector feedback forces. By remotely operating the buttons on the master arm, the operating mode of the master arm can be switched between a single-arm posture decoupling remote operation mode and a dual-arm collaborative posture decoupling remote operation mode.
[0043] Example 2
[0044] This embodiment provides a dual-arm pose decoupling teleoperation method for single-master remote control of dual-slave arms, including:
[0045] S201: The main controller receives instructions for the material handling operation target;
[0046] S202: The master controller performs a decoupled teleoperation of the left slave robot arm's posture, sending the teleoperation command to the left slave robot arm controller. The left slave robot arm controller controls the left arm to adjust its position according to the teleoperation command, so that the left arm reaches the vicinity of the work target. The left slave robot arm controller adjusts the posture of the left arm according to the teleoperation command, so that the left arm grasps the left end of the work target.
[0047] The master controller performs a decoupled teleoperation of the right slave robot arm's posture, sending teleoperation commands to the right slave robot arm controller. The right slave robot arm controller controls the right arm to adjust its position according to the teleoperation commands, so that the right arm reaches the vicinity of the work target. The right slave robot arm controller adjusts the posture of the right arm according to the teleoperation commands, so that the right arm can grasp the right end of the work target.
[0048] S203: The master controller performs collaborative pose decoupling teleoperation of the two slave robotic arms, and sends teleoperation commands to the left slave robotic arm controller and the right slave robotic arm controller. The left slave robotic arm controller and the right slave robotic arm controller control the pose and joint angles of the left slave robotic arm and the right slave robotic arm respectively, so as to achieve the adjustment of the position and posture of the work target.
[0049] S204: The master controller is based on admittance control to achieve active compliance in the center position of the work target.
[0050] Furthermore, in S202: the master controller performs the left slave robot pose decoupling teleoperation, which is consistent with the specific steps of the master controller performing the right slave robot pose decoupling teleoperation.
[0051] Figure 4 This is the overall flowchart of the pose decoupling teleoperation method for single master-hand remote control of two slave arms mentioned in this invention. In the dual-arm collaborative pose decoupling teleoperation process, this invention uses a single-end teleoperation master hand to remotely control two slave-end robotic arms. First, the teleoperation master hand is set to operate in the left arm pose decoupling teleoperation mode, as... Figure 5 The method described in the article, which involves telemanipulating the left arm through pose decoupling, first adjusts the position of the left arm to bring it around the target area. Then, it adjusts the posture of the left arm to grasp the left end of the target. Next, the telemanipulation mode of the master hand is switched to the right arm pose decoupling telemanipulation mode. Similarly, pose decoupling is used to control the position and posture of the right arm's end cap, allowing the right arm to grasp the right end of the target.
[0052] Figure 5This is a schematic diagram of the workflow when the teleoperated master hand is in a single-arm pose-decoupled teleoperation state. The first, second, and third joints of the master hand adopt a linkage structure, while the fourth, fifth, and sixth joints are integrated into the end effector sphere. The first, second, and third joints apply the position mode teleoperation concept, while the fourth, fifth, and sixth joints apply the rate mode concept. The relative change of the center position of the sphere is mapped to the relative change of the end effector position of the robotic arm. The presence of a button can change the mapping scaling factor K. The fourth, fifth, and sixth joints control the rotation of the end effector coordinate system around the X-axis, Y-axis, and Z-axis, respectively. The three axes of the fourth, fifth, and sixth joints control the rotation of the end effector coordinate system around the X-axis, Y-axis, and Z-axis.
[0053] Further, S202: The master controller performs a left slave robot pose decoupling teleoperation, including:
[0054] The angle information of the six joints of the single master arm is known. Using the real-time angle information of the first, second, and third joints, the real-time position of the ball device at the end of the single master arm is calculated using the positive kinematics of the first three joints. The calculation formula is as follows:
[0055]
[0056] Where p x p y p z θ1, θ2, and θ3 represent the positions of the end effector ball device of the single master arm in the X, Y, and Z axes of the single master arm base coordinate system, respectively, and represent the angles of its first, second, and third joints.
[0057] The end-effector position of the left or right slave robotic arm is obtained through position mapping; assuming the motion space of the teleoperated master hand is Ω. m The motion space of the robotic arm is Ω s Then let:
[0058]
[0059]
[0060] Among them, a m b represents the maximum distance of the teleoperator's workspace in the X-axis direction. m c represents the maximum distance of the teleoperator's workspace in the Y-axis direction. m The maximum distance of the teleoperator's workspace in the Z-axis direction is represented by p. mxmax p represents the maximum value of the teleoperated master's workspace along the X-axis. mxminp represents the minimum value of the teleoperation master's workspace along the X-axis. mymax p represents the maximum value of the teleoperation master's workspace along the Y-axis. mymin p represents the minimum value of the teleoperation master's workspace along the Y-axis. mzmax p represents the maximum value of the teleoperator's workspace along the Z-axis. mzmin This represents the minimum value of the teleoperated master's workspace along the Z-axis; a s b represents the maximum distance from the workspace of the robotic arm in the X-axis direction. s c represents the maximum distance from the workspace of the robotic arm in the Y-axis direction. s p represents the maximum distance from the workspace of the robotic arm in the Z-axis direction. sxmax p represents the maximum value of the robotic arm's workspace in the X-axis direction. sxmin p represents the minimum value of the robotic arm's workspace in the X-axis direction. symax p represents the maximum value of the robotic arm's workspace in the Y-axis direction. symin p represents the minimum value of the robotic arm's workspace in the Y-axis direction. szmax p represents the maximum value of the robotic arm's workspace in the Z-axis direction. szmin This represents the minimum value of the workspace of the robotic arm in the Z-axis direction.
[0061] Collect the joint angle information of the master arm, and obtain the desired position of the slave arm through mapping:
[0062]
[0063] Among them, P s (t) represents the desired position of the end effector of the robotic arm at time t, p sx (t) represents the desired position of the end effector of the robotic arm in the X-axis direction at time t, p sy (t) represents the desired position of the end effector of the robotic arm in the Y-axis direction at time t, p sz (t) represents the desired position of the end effector of the robotic arm in the Z-axis direction at time t, p mx (t) represents the position of the end effector sphere of the single-master robotic arm along the X-axis at time t, p my (t) represents the position of the end effector sphere of the single-master robotic arm along the Y-axis at time t, p mz (t) represents the position of the end effector ball of the single master arm in the Z-axis direction at time t;
[0064] When the end-effector of the single-master robotic arm changes position, the master controller acquires the position changes Δθ4, Δθ5, and Δθ6 of the fourth, fifth, and sixth joints in real time.
[0065] We consider the position changes of the fourth, fifth, and sixth joints as rotations of the end-effector coordinate system around the X, Y, and Z axes, respectively. The position changes of these three joints can generate three attitude change matrices R for the end-effector posture. x (Δθ4), R y (Δθ5), R z (Δθ6) can be calculated as follows:
[0066]
[0067] k in the formula i It is a variable positive parameter, so the desired end effector posture matrix of the slave robot at time t can be obtained:
[0068] R s (t)=R x (Δθ4)R y (Δθ5)R z (Δθ6)R s (t-1)
[0069] Therefore, the desired pose matrix T of the slave robot at any time is... s (t), represented as:
[0070]
[0071] Among them, R s (t) represents the desired end-effector posture matrix at time t, P s (t) represents the desired position of the slave robot arm at time t.
[0072] Finally, by solving the inverse kinematics of the slave arm, the angles of each joint of the slave arm are obtained.
[0073] It should be understood that the above steps can ensure the transparency of remote operation while also guaranteeing a certain range of motion capabilities and precise positioning within a small area of space, thereby enhancing the user's operating experience.
[0074] After a single teleoperated master hand uses pose-decoupled teleoperation to allow both arms to grasp the two ends of the target, the teleoperated master hand's working mode is changed to a dual-arm collaborative pose-decoupled teleoperation mode, such as... Figure 6 The method described in the text. At this point, both arms have grasped the target to be worked on, and the control variables become the position and attitude of the target. The position and attitude of the target are changed separately by teleoperating the master hand. Figure 3The strong constraints described can be used to obtain the angles of each joint in both arms and determine whether they meet the maximum allowable deviation. If they are within the maximum allowable deviation, the servo angles can be sent to the controller.
[0075] Figure 6 This is a flowchart illustrating pose decoupling telemanipulation of a single telemanipulator master hand to two slave arms. The pose of the workpiece is used as the control variable for the master hand. The position of the workpiece is mapped to the position of the workpiece using the end effector position of the third joint of the master hand, and the changes in the pose of the workpiece are mapped to the changes in the pose of the workpiece using the changes in the fourth, fifth, and sixth joints of the master hand. The specific mapping method and process are detailed below. Figure 4 Consistent with. According to... Figure 3 The strong constraint conditions introduced in the paper for cooperative motion allow the poses of both arms to be uniquely determined when the pose of the cooperative task target is known. This enables the use of a master arm to decouple the poses of two slave arms to complete a cooperative task.
[0076] Further, S203: The master controller performs a dual-slave robotic arm cooperative pose decoupling teleoperation, including:
[0077] Calculate the desired pose matrix of the task target; calculate the desired pose matrix of the two arms through the pose constraints of the coordinated movement of the two arms; calculate the maximum position deviation constraint and analyze whether the movement of the two arms meets the requirements;
[0078] The desired pose matrix of the computational task target includes:
[0079] The angle information of the six joints of the single master arm is known. Using the real-time angle information of the first, second, and third joints, the real-time position of the end effector ball device of the single master arm is calculated:
[0080]
[0081] Where p x p y p z θ1, θ2, and θ3 represent the positions of the end effector ball device of the single master arm in the X, Y, and Z axes of the single master arm base coordinate system, respectively, and represent the angles of its first, second, and third joints.
[0082] The target position is obtained through position mapping. We have previously assumed that the motion space of the teleoperator is Ω. m Let's assume the motion space of the target is Ω. s2 Then let:
[0083]
[0084]
[0085] Among them, a m b represents the maximum distance of the teleoperator's workspace in the X-axis direction. m c represents the maximum distance of the teleoperator's workspace in the Y-axis direction. m The maximum distance of the teleoperator's workspace in the Z-axis direction is represented by p. mxmax p represents the maximum value of the teleoperated master's workspace along the X-axis. mxmin p represents the minimum value of the teleoperation master's workspace along the X-axis. mymax p represents the maximum value of the teleoperation master's workspace along the Y-axis. mymin p represents the minimum value of the teleoperation master's workspace along the Y-axis. mzmax p represents the maximum value of the teleoperator's workspace along the Z-axis. mzmin This represents the minimum value of the teleoperated master's workspace along the Z-axis; a s2 b represents the maximum distance in the X-axis direction of the target's movement space. s2 c represents the maximum distance in the Y-axis direction of the target's movement space. s2 p represents the maximum distance in the Z-axis direction of the target's motion space. s2xmax p represents the maximum value of the target's motion space in the X-axis direction. s2xmin p represents the minimum value of the motion space of the target object in the X-axis direction. s2ymax p represents the maximum value of the target's motion space in the Y-axis direction. s2ymin p represents the minimum value of the motion space of the target object in the Y-axis direction. s2zmax p represents the maximum value of the target's motion space in the Z-axis direction. s2zmin This represents the minimum value of the movement space of the target object in the Z-axis direction.
[0086] Collect the joint angle information of the dominant hand, and obtain the desired position of the task target through mapping:
[0087]
[0088] P s2 (t) represents the desired position of the task objective at time t, p s2x (t) represents the desired position of the target object in the X-axis direction at time t, p s2y (t) represents the desired position of the target object in the Y-axis direction at time t, p s2z (t) represents the desired position of the target object in the Z-axis direction at time t, p mx (t) represents the position of the single-master robotic arm along the X-axis at time t, p my(t) represents the position of the single-master robotic arm in the Y-axis direction at time t, p mz (t) represents the position of the single master arm in the Z-axis direction at time t;
[0089] When the end-effector of the single-master robotic arm changes position, the master controller acquires the position changes Δθ4, Δθ5, and Δθ6 of the fourth, fifth, and sixth joints in real time.
[0090] The positional changes of the fourth, fifth, and sixth joints can be considered as rotations of the target coordinate system around the X, Y, and Z axes, respectively. These three joint positional changes can generate three attitude change matrices R for the target's posture. x (Δθ4), R y (Δθ5), R z (Δθ6) can be calculated as follows:
[0091]
[0092] Where, k i It is a variable positive parameter, therefore the expected attitude matrix of the target at time t can be obtained:
[0093] R s2 (t)=R x (Δθ4)R y (Δθ5)R z (Δθ6)R s2 (t-1)
[0094] Therefore, the expected pose matrix T of the target at any time is... s (t), represented as:
[0095]
[0096] Among them, R s2 (t) represents the desired end-effector posture matrix at time t, P s2 (t) represents the desired position of the slave robot arm at time t.
[0097] Further, the calculation of the desired pose matrix of the two arms through the pose constraint of the coordinated movement of the two arms includes:
[0098] The coordinate system of the centroid of the target to be worked on is transformed into the dual-arm base coordinate system through the left and right arm base coordinate systems:
[0099]
[0100] in, b T o The matrix representing the pose and motion description of the target in the base coordinate system;b R ol and b R or The homogeneous coordinate transformation matrix of the left and right robotic arms' bases in the base coordinate system represents the orientation of the bases. ol R 6l and or R 6r This represents the attitude transformation matrix of the left and right robotic arms at their respective base coordinate systems. 6l R o and 6r R o This represents the attitude transformation matrix of the working target of the left and right robotic arms in their respective end-effector coordinate systems. b P ol and b P or The homogeneous coordinate transformation matrix represents the position of the base of the left and right robotic arms in the base coordinate system; b P 6l and b P 6r This represents the homogeneous coordinate transformation matrix of the pose positions of the left and right ends of the robotic arms in the base coordinate system.
[0101] In terms of position, the position terms of the pose matrices of the joint operation target in the coordinate systems of the two hydraulic arm bases can be obtained as follows:
[0102] b R ol ol R 6l 6l R o1 + b P ol + b R ol b P 6l = b R or or R 6r 6r R o2 + b P or + b R or b P 6r ;
[0103] After solving and simplifying, the positional relationship between the homogeneous pose transformation matrices of the ends of the two arms corresponding to their respective bases is as follows:
[0104] or P 6r =k1(k2+k3( ol R 6l k4+ol P 6l ));
[0105] in, k2 = b P ol - b P or k3 = b R ol k4 = 6l R0( o P 6r - o R 6l ).
[0106] In terms of attitude, by simultaneously determining the attitude terms of the pose matrices of the target in the coordinate systems of the two hydraulic arm bases, we can obtain:
[0107] b R ol ol R 6l 6l R o1 = b R or or R 6r 6r R o2
[0108] After simplification, the attitude relationship between the homogeneous pose transformation matrices of the ends of the two arms corresponding to their respective bases can be obtained as follows:
[0109] or R 6r =k1k3 ol R 6l k5
[0110] Where, k5 = 6l R o ( 6r R o ) -1
[0111] Next, based on the position constraints, we will consider velocity constraints and acceleration constraints.
[0112] Position constraints on the task target b P 6l (q l )= b P 6r (q r Taking the derivative with respect to time, we get:
[0113]
[0114] Among them, Jλ (q l ) represents the Jacobian matrix of the left arm. J represents the velocity of the left arm joint. λ (q r ) represents the Jacobian matrix of the right arm. Indicates the speed of the right arm joint;
[0115] When two arms work together to move a rigid body, the linear velocity of the rigid body is constant, and the linear velocities of the end effectors of the two arms are always the same.
[0116] The angular velocity constraint of the dual-arm end effector can be obtained by differentiating the position constraint of the dual-arm end effector. However, since there is no relative motion between the two end effectors when transporting a rigid body, their angular velocities are exactly the same. Therefore, the angular velocity constraint can be directly derived from the angular velocity.
[0117]
[0118] Among them, J λ (q l ) represents the Jacobian matrix of the left arm. J represents the velocity of the left arm joint. λ (q r ) represents the Jacobian matrix of the right arm. Indicates the speed of the right arm joint;
[0119] By combining the equations, we can deduce that the relationship between the velocities of the joints in both arms should be:
[0120]
[0121] in,
[0122] Further targeting Taking the derivative over time, we can find the constraints on the acceleration of the joints in both arms:
[0123]
[0124] in,
[0125] From the constraints on the acceleration of the two arm joints, it can be seen that... From third-order tensors and Because it involves four related factors, it is non-linear with respect to the joint velocities and accelerations of both arms. In actual operational scenarios, the velocity and acceleration of one arm can be obtained from sensors, so the velocity and acceleration of the other arm can be calculated from the velocity and acceleration of the first arm.
[0126] Furthermore, the constraints on the maximum positional deviation include:
[0127] remember r T l Let be the homogeneous transformation matrix from the base coordinate system of the left arm to the base coordinate system of the right arm. 6r T 6l The relative transformation matrix from the end effector coordinate system of the left arm to the end effector coordinate system of the right arm. l1 T l2 and r1 T r2 Let be the relative transformation matrices of the two arms from time t1 to time t2. Then, during the process from time t1 to time t2, the two arms need to satisfy the following constraints:
[0128] l1 T l2 6r T 6l = 6r T 6l r1 T r2
[0129] When the relative displacement of the end effectors of the dual-slave robotic arm is less than a set threshold, l1 T l2 and r1 T r2 It can be considered as l1 T l2 =δ( l1 T l2 ), r1 T r2 =δ( r1 T r2 ),have:
[0130]
[0131]
[0132] Here, δ is the mathematical symbol for differential, representing a differential transformation;
[0133] right l1 T l2 6r T 6l = 6r T 6l r1 T r2 According to the theory of differential transformation, we can obtain:
[0134] δp l =J lr δp r ;
[0135] δp l =[d lx d ly d lz δ lx δ ly δ lz ];
[0136] δp r =[d rx d ry d rz δ rx δ ry δ rz ];
[0137]
[0138]
[0139] And there are position and orientation constraints:
[0140] T l 6r T 6l = r T l T r ;
[0141] During the motion from time t1 to t2, let the point calculated by the trajectory of the two arms be {X}. li} and {X ri} where (i = 0, 1, 2, ..., n), then we have:
[0142] T li 6r T 6l = r T l T ri ;
[0143] In the formula,
[0144] Substituting it into the expansion yields:
[0145] R li 6r P 6l +P li =R x P ri +P x ;
[0146] The positional deviation of the two arms is defined as:
[0147] δ p (t)=||R li (t) 6r P6l +P li (t)-R x P ri (t)-P x ||
[0148] Should meet Given the maximum allowable error, the maximum positional deviation constraint should be:
[0149]
[0150] If the inverse kinematics processing of the desired pose matrix of the two arms yields joint angles that satisfy the maximum position deviation constraint, then the solved joint angles are effectively programmable.
[0151] It should be understood that Figure 3 It is a process of coordinated movement of both arms. In this case, the end effectors of both arms and the target to be worked on are relatively stationary, and the three form a closed kinematic chain.
[0152] Furthermore, S204: The master controller, based on admittance control, achieves active compliance of the target center position, including:
[0153] When the end effector is in operation, it treats the target object as a damped model with virtual springs at both ends. Since the coordinated motion of the entire double-arm transport rigid body is a completely closed chain, when one end of the working rigid body is subjected to a force, the other end is subjected to a force of equal magnitude and opposite direction. Through admittance control design based on the center position of the target object, active compliance is achieved to prevent damage to the target rigid body due to excessive end force.
[0154] Furthermore, S204: The master controller, based on admittance control, achieves active compliance of the target center position, including:
[0155]
[0156] Among them, F l and F r K represents the contact force measured by the six-dimensional force sensors at the ends of the left and right arms, respectively. d B d M d The desired parameters of the set admittance controller are represented by X, which are six-dimensional stiffness diagonal matrices, six-dimensional damping diagonal matrices, and six-dimensional mass diagonal matrices, respectively. r , Let X represent the pose, velocity, and acceleration of the reference target, respectively, which are known quantities. It is the desired pose of the task objective.
[0157] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A dual-arm pose decoupled teleoperation method of a single master remote controlling dual slave arms, characterized in that, include: It features a single master robotic arm and dual slave robotic arms; A dual-slave robotic arm includes a left slave robotic arm and a right slave robotic arm. The single master-hand robotic arm includes a first base, on which a first joint is mounted, and on which a second joint is mounted. The second joint is connected to a third joint via a first link, and the third joint is connected to an end effector via a second link. The end effector includes a first columnar body, on which a fourth joint is mounted, and on which a ball is mounted. A fifth joint is mounted inside the ball, and the fifth joint is connected to a sixth joint via a third link. The master-hand controller receives instructions for handling the target object. The master controller performs a decoupled teleoperation of the left slave robot arm's posture, sending teleoperation commands to the left slave robot arm controller. The left slave robot arm controller controls the left slave robot arm to adjust its position according to the teleoperation commands, so that the left slave robot arm reaches the vicinity of the work target. The left slave robot arm controller adjusts the posture of the left slave robot arm according to the teleoperation commands, so that the left slave robot arm can grasp the left end of the work target. The master controller performs a remote operation to decouple the pose of the left slave robotic arm, including: The angle information of the six joints of the single master arm is known. The real-time position of the ball device at the end of the single master arm is calculated based on the positive kinematics of the first three joints using the real-time angle information of the first, second and third joints. The end position of the left slave robot arm is obtained through position mapping; Collect the joint angle information of the single master arm robot and obtain the desired position of the slave arm robot through mapping. When the end sphere device of the single master hand mechanical arm changes in posture, the master controller acquires the position change amount of the fourth joint, the fifth joint and the sixth joint in real time , , ; The position changes of the fourth joint, the fifth joint and the sixth joint are respectively regarded as rotations of the end-effector coordinate system of the slave manipulator around the X axis, the Y axis and the Z axis respectively; the position changes of the three joints respectively generate three attitude change matrices of the end-effector attitude of the slave manipulator 、 、 ; Therefore, the desired end-effector pose matrix of the slave robot at time t is obtained; and the desired pose matrix of the slave robot at any time is then obtained. Finally, by solving the inverse kinematics of the slave arm, the angles of each joint of the slave arm are obtained. The master controller performs a decoupled teleoperation of the right slave robot arm's posture, sending teleoperation commands to the right slave robot arm controller. The right slave robot arm controller controls the right slave robot arm to adjust its position according to the teleoperation commands, so that the right slave robot arm reaches the vicinity of the work target. The right slave robot arm controller adjusts the posture of the right slave robot arm according to the teleoperation commands, so that the right slave robot arm can grasp the right end of the work target. The master controller performs collaborative pose decoupling teleoperation of the two slave robotic arms, sending teleoperation commands to the left slave robotic arm controller and the right slave robotic arm controller. The left slave robotic arm controller and the right slave robotic arm controller control the pose and joint angles of the left and right slave robotic arms respectively, so as to achieve the adjustment of the position and posture of the work target. The master controller is based on admittance control to achieve active compliance in the center position of the work target.
2. The dual-arm pose decoupling teleoperation method for single master-hand remote control of dual slave arms as described in claim 1, characterized in that, The master controller performs collaborative pose decoupling teleoperations between the two slave robotic arms, including: Calculate the desired pose matrix of the task target; calculate the desired pose matrix of the two arms through the pose constraints of the coordinated movement of the two arms; calculate the maximum position deviation constraint and analyze whether the movement of the two arms meets the requirements; If the joint angles of the two arms are obtained by performing inverse kinematics processing on the expected pose matrix of the two arms, and the maximum position deviation constraint is satisfied, then the joint angles obtained are valid.
3. The dual-arm pose decoupling teleoperation method for single master-hand remote control of dual slave arms as described in claim 2, characterized in that, The calculation of the desired pose matrix of the two arms through the pose constraint of the coordinated movement of the two arms includes: The coordinate system of the centroid of the target is transformed into the dual-arm base coordinate system through the coordinate systems of the left and right slave robotic arms. In terms of position, the position terms of the pose matrices of the joint operation target in the coordinate systems of the left and right robotic arms bases are solved to obtain the positional relationship between the homogeneous pose transformation matrices of the ends of the two arms corresponding to their respective bases. In terms of attitude, the attitude terms of the pose matrices of the target in the coordinate systems of the left and right robotic arms are combined to obtain the attitude relationship between the homogeneous pose transformation matrices of the ends of the two arms corresponding to their respective bases. Based on the position constraints, consider the velocity and acceleration constraints; differentiate the position constraints of the task target with respect to time; derive the angular velocity constraints directly from the angular velocity; and find the constraint conditions for the acceleration of the two arm joints.
4. The dual-arm pose decoupling teleoperation method for single master-hand remote control of dual slave arms as described in claim 1, characterized in that, The master controller, based on admittance control, achieves active compliance in positioning the target center, including: When the end effector is in operation, it treats the target as a damped model with virtual springs at both ends. Since the coordinated motion of the entire double-arm carrying the rigid body of the target is a complete closed chain, when one end of the target is subjected to a force, the other end is subjected to a force of equal magnitude and opposite direction. Through admittance control design based on the center position of the target, active compliance is achieved to prevent damage to the target due to excessive end force.
5. The dual-arm pose decoupling teleoperation method for single master-hand remote control of dual slave arms as described in claim 1, characterized in that, The master controller, based on admittance control, achieves active compliance at the center position of the work target, including: in, and These represent the contact forces measured by the six-dimensional force sensors at the ends of the left and right robotic arms, respectively. , , The desired parameters for the set admittance controller are represented by the six-dimensional stiffness diagonal matrix, the six-dimensional damping diagonal matrix, and the six-dimensional mass diagonal matrix, respectively. , , Let these represent the pose, velocity, and acceleration of the target object being referenced, respectively; these are known quantities. , , It is the desired pose of the task objective.
6. The dual-arm pose decoupling teleoperation method for single master-hand remote control of dual slave arms as described in claim 1, characterized in that, The master controller performs the left slave robot pose decoupling teleoperation, and the specific steps are the same as those of the master controller performing the right slave robot pose decoupling teleoperation.
7. A system for a dual-arm pose decoupling teleoperation method with single master hand remote control of dual slave arms as described in any one of claims 1-6, characterized in that, The single master robotic arm is equipped with a master controller. The master controller performs decoupled teleoperation of the left or right slave robotic arm, sending teleoperation commands to the left or right slave robotic arm controller. The left or right slave robotic arm controller controls the left slave robotic arm to adjust its position according to the teleoperation commands, so that the left or right slave robotic arm reaches the vicinity of the work target. The left or right slave robotic arm controller adjusts the posture of the left or right slave robotic arm according to the teleoperation commands, so that the left or right slave robotic arm can grasp the end of the work target. The master controller performs collaborative pose decoupling teleoperation of the two slave robotic arms, sending teleoperation commands to the left slave robotic arm controller and the right slave robotic arm controller. The left slave robotic arm controller and the right slave robotic arm controller control the pose and joint angles of the left and right slave robotic arms respectively, so as to achieve the adjustment of the position and posture of the work target. The master controller is based on admittance control to achieve active compliance in the center position of the work target.
8. The system of a dual-arm pose decoupling teleoperation method for single master-hand remote control of dual slave arms as described in claim 7, characterized in that, The left and right robotic arms have the same structure. The left robotic arm includes a second base connected in sequence. The second base is connected to the end effector through several sets of joint and link assemblies. Each set of joint and link assemblies includes joints and links connected in sequence.
9. The system of a dual-arm pose decoupling teleoperation method for single master-hand remote control of dual slave arms as described in claim 7, characterized in that, The main hand controller is connected to the main hand encoder, motor and buttons respectively; The master controller is connected to the left slave robot controller and the right slave robot controller via a network. The left-hand robotic arm controller is also connected to the left-hand encoder, the left-hand six-dimensional force sensor, and the left-hand voltage-current converter; the left-hand voltage-current converter is connected to the left-hand hydraulic cylinder through the left-hand electro-hydraulic servo valve; the left-hand hydraulic cylinder is connected to the left-hand robotic arm body; the left-hand encoder and the left-hand six-dimensional force sensor are connected to the left-hand robotic arm body. The right-hand robotic arm controller is also connected to the right-hand encoder, the right-hand six-dimensional force sensor, and the right-hand voltage-current converter; the right-hand voltage-current converter is connected to the right-hand hydraulic cylinder through the right-hand electro-hydraulic servo valve; the right-hand hydraulic cylinder is connected to the right-hand robotic arm body; the right-hand encoder and the right-hand six-dimensional force sensor are connected to the right-hand robotic arm body.