Whole body control method for mobile hybrid robot

Abstract

The application provides a full-body control method of a mobile compound robot, comprising the following steps: S1, establishing an integrated system model in a world coordinate system; S2, analyzing external data instructions into input control quantities of a full-body controller; S3, transmitting state parameters to the integrated system model and transmitting sensing data to the full-body controller; S4, transmitting system theoretical adjustment vector parameters to the full-body controller; S5, generating total control quantities; S6, performing inverse analysis calculation to obtain control quantities of a mobile body and operation control quantities of a mechanical arm; and S7, transmitting the control quantities to corresponding controllers of the mobile body and the mechanical arm respectively. The application does not separately distinguish how to calculate the control quantities of the mobile body and the mechanical arm of the robot, and only after the total control quantities are finally obtained, the mobile body movement control quantities and the mechanical arm control quantities are analyzed based on the total control quantities, so that the walking and operation actions of the robot have strong synchronism and coordination.

Description

Technical Field

[0001] This invention relates to the field of mobile composite robot control technology, and in particular to a method for full-body control of a mobile composite robot. Background Technology

[0002] Mobile composite robots consist of a mobile chassis and a robotic arm. Traditional mobile composite robot control often employs a separate sequential control approach. That is, the motion control of the robot's mobile chassis and the motion control of the robotic arm are decoupled and separate; the control behavior is also sequential. For example, the robot's mobile chassis moves to the target position first and then stops; subsequently, the robotic arm begins to perform the operation. Due to the limitations of this control method, achieving synchronized and coordinated movements between the mobile chassis and the robotic arm is quite difficult.

[0003] This control strategy results in poor flexibility in robot movement and operation, making it difficult to synchronize actions and leading to low overall operational efficiency. It is not suitable for some high-speed operation scenarios.

[0004] Therefore, a control method is needed that can enable robot movement, more flexible robotic arm movements, and synchronous and coordinated control effects to meet the requirements of high-performance robots. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention discloses a method for full-body control of a mobile composite robot. The technical solution of this invention is implemented as follows:

[0006] A method for controlling the whole body of a mobile composite robot includes the following steps:

[0007] S1, Establish an integrated system model of the mobile composite robot in the world coordinate system;

[0008] S2, after receiving external data instructions, the robot controller parses the external data instructions into input control quantities for the whole-body controller and transmits them to the whole-body controller and the integrated system model;

[0009] S3, the mobile robot transmits the state parameters of the mobile composite robot to the integrated system model; the mobile composite robot transmits the perception data of the mobile composite robot to the whole-body controller;

[0010] S4, the integrated system model transmits the generated system theoretical adjustment vector parameters to the whole-body controller;

[0011] S5, the whole-body controller generates the total control quantity based on the input control quantity, the system theory, the adjusted vector parameters, and the robot's perception data;

[0012] S6, the whole-body controller performs inverse analytical calculations to obtain the control quantities of the mobile body of the mobile composite robot and the operation control quantities of the robotic arm, respectively;

[0013] S7, the whole body controller transmits the control quantities of the mobile body of the mobile composite robot and the operation control quantities of the robotic arm to the hub driver corresponding to the mobile body and the servo joint driver corresponding to the robotic arm, respectively.

[0014] In this invention, perception data refers to the data from the various sensors carried by the robot. After the robot's system processes and fuses these data, it forms the robot's perception capability of the external environment.

[0015] Preferably, in step S5, the whole-body controller establishes a multi-task priority null space projection method based on the input control quantity, the system theoretical adjustment vector parameters, and the sensing data. The low-priority tasks are placed in the null space of the high-priority tasks for calculation, and then the control quantities of all tasks are added together to obtain the total control quantity.

[0016] Preferably, step S5 further includes a dynamic weight optimization strategy;

[0017] The whole-body controller evaluates the operation results based on the historical index parameter data of the mobile composite robot, and determines whether to optimize the index parameter data corresponding to the operation results based on the evaluation results, and adds the dynamically optimized weight parameters to the generation of the total control quantity.

[0018] Preferably, in step S1, the integrated system model couples the mobile body mathematical model of the mobile composite robot and the robotic arm mathematical model;

[0019] The mathematical model of the mobile body of the mobile composite robot is as follows:

[0020]

[0021] Where A and B are parameter matrices related to the mechanical structure of the robot body;

[0022] The vector of rotational speed of the robot's left and right drive wheels;

[0023] The angular velocity of the moving body.

[0024] The derivative of the position parameters of the robotic arm base on the moving body;

[0025] The mathematical model of the robotic arm is:

[0026] -f0

[0027] Where, r eThis is the pose vector of the robotic arm's end effector;

[0028] f(θ) represents a function that can be expressed using the known structural parameters of the robotic arm and is used to describe the pose behavior of the robotic arm's end effector.

[0029] The integrated system model of the mobile composite robot is as follows:

[0030]

[0031] in, This represents the velocity vector matrix at the end effector of the mobile composite robot arm.

[0032] The Jacobian matrix represents the simultaneous coupling of the moving body's pose and the robotic arm's end effector's pose angle.

[0033] k is the parameter matrix of the mobile robot;

[0034] θ is the pose angle of the robotic arm;

[0035] This represents the velocity vector of the robotic arm's motion pose.

[0036] Preferably, the mathematical expression for the whole-body controller is:

[0037]

[0038] Where τ is the planning torque weighted by the weights;

[0039] ω1, ω2, ..., ω k Dynamically optimize the weighting coefficients for tasks corresponding to different historical indicator parameter data;

[0040] J1, J2, ..., J k Jacobian matrix for tasks corresponding to different historical indicator parameters;

[0041] F1, F2, ..., F k This represents the Cartesian impedance space matrix corresponding to different historical index parameter data.

[0042] Preferably, the state parameters of the mobile composite robot include the current walking speed, angle, acceleration, and robotic arm motion parameters.

[0043] Preferably, the historical performance parameters of the mobile composite robot include operation cycle time, operation duration, operation error, and stability of the robotic arm end effector.

[0044] Preferably, the evaluation results are graded as excellent, good, or average.

[0045] The indicator parameters corresponding to operation results rated as excellent will no longer be optimized, the indicator parameters corresponding to operation results rated as good will be optimized at the second level, and the indicator parameters corresponding to operation results rated as average will be optimized at the first level.

[0046] In this invention, first-level optimization and second-level optimization are used only to indicate the degree of optimization. Generally, the lower the evaluation result level, the worse the corresponding operation result, and therefore, a higher degree of optimization is required. In this invention, the degree of first-level optimization is higher than that of second-level optimization.

[0047] This invention views the entire system from the perspective of the robotic arm of a mobile composite robot, essentially considering it as a system with redundant degrees of freedom. During operation, this system performs multiple tasks while also being subject to multiple constraints. This invention employs integrated whole-body control, utilizing the system's redundancy to establish a multi-task framework model. Based on data commands such as robot trajectory instructions and operation commands, a priority null-space projection method is established for each task. Low-priority tasks are calculated within the null space of high-priority tasks. The control quantities of all tasks are then summed to obtain control quantities with different priorities, allowing the control quantities to complete low-priority tasks without violating high-priority tasks. To further optimize tracking accuracy and balance the control effects of each task, a dynamic weight optimization mechanism is established to achieve collaborative control. This enables the robot to synchronously control the robotic arm's movements while moving, and ensures that the robot arm's end effector achieves both correct angular orientation and the required operational accuracy during final operation. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only one embodiment of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of the parameters of a mobile composite robot based on the world coordinate system, as an embodiment of a whole-body control method for a mobile composite robot.

[0050] Figure 2 This is a schematic diagram of the control system structure of an integrated system model in an embodiment of a whole-body control method for a mobile composite robot.

[0051] Figure 3 A reverse analytical flowchart of the control system in an embodiment of a whole-body control method for a mobile composite robot;

[0052] Figure 4This is a flowchart illustrating the dynamic optimization and adjustment of weight parameters in an embodiment of a whole-body control method for a mobile composite robot. Detailed Implementation

[0053] 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 embodiments of the present invention, and not all embodiments. 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.

[0054] Example

[0055] In one specific embodiment, a method for whole-body control of a mobile composite robot includes the following steps:

[0056] S1, as Figure 1 As shown, an integrated system model of the mobile composite robot is established in the world coordinate system;

[0057] The mathematical model of the mobile body of the mobile composite robot is as follows:

[0058]

[0059] Where A and B are parameter matrices related to the mechanical structure of the robot body;

[0060] The vector of rotational speed of the robot's left and right drive wheels;

[0061] The angular velocity of the moving body.

[0062] The derivative of the position parameters of the robotic arm base on the moving body;

[0063] In this embodiment, the location of the end effector of the robotic arm is taken as the final object of consideration, and the robotic arm is regarded as a system with redundant degrees of freedom. Its mathematical model can be expressed by the following expression:

[0064] re = f(θ)

[0065] Where, r e This is the pose vector of the robotic arm's end effector;

[0066] f(θ) represents a function that can be expressed using the known structural parameters of the robotic arm and is used to describe the pose behavior of the robotic arm's end effector.

[0067] In this embodiment, the integrated system model of the mobile composite robot, which simultaneously couples the mobile body and the robotic arm model in the world coordinate system, is as follows:

[0068]

[0069] in, This represents the velocity vector matrix at the end effector of the mobile composite robot arm.

[0070] The Jacobian matrix represents the simultaneous coupling of the moving body's pose and the robotic arm's end effector's pose angle.

[0071] k is the parameter matrix of the mobile robot;

[0072] θ is the pose angle of the robotic arm;

[0073] This represents the velocity vector of the robotic arm's motion pose.

[0074] like Figure 2 As shown, this embodiment, based on the establishment of an integrated system model, adopts a control strategy based on an integrated whole-body control method combined with zero-space projection, and uses a dynamic variable weight method to dynamically adjust the priority of each task, thereby improving the collaborative control effect of the mobile composite robot.

[0075] S2, after receiving external data instructions, the robot controller parses the external data instructions into input control quantities for the whole-body controller and transmits them to the whole-body controller and the integrated system model;

[0076] The mathematical expression for the whole-body controller is:

[0077]

[0078] Where τ is the planning torque weighted by the weights;

[0079] ω1, ω2, ..., ω k Dynamically optimize the weighting coefficients for tasks corresponding to different historical indicator parameter data;

[0080] J1, J2, ..., J k Jacobian matrix for tasks corresponding to different historical indicator parameters;

[0081] F1, F2, ..., F k This represents the Cartesian impedance space matrix corresponding to different historical index parameter data.

[0082] S3, the mobile robot transmits the state parameters of the mobile composite robot to the integrated system model; the mobile composite robot transmits the perception data of the mobile composite robot to the whole-body controller;

[0083] S4, the integrated system model transmits the generated system theoretical adjustment vector parameters to the whole-body controller;

[0084] S5, the whole body controller establishes a multi-task priority null space projection method based on input control quantity, system theoretical adjustment vector parameters and sensing data. It places low-priority tasks in the null space of high-priority tasks for calculation, and then adds the control quantities of all tasks to obtain the total control quantity.

[0085] S6, such as Figure 3 As shown, the whole-body controller performs inverse analytical calculations to obtain the control quantities of the mobile body of the mobile composite robot and the operation control quantities of the robotic arm, respectively.

[0086] S7, the whole body controller transmits the control quantities of the mobile body of the mobile composite robot and the operation control quantities of the robotic arm to the hub driver corresponding to the mobile body and the servo joint driver corresponding to the robotic arm, respectively.

[0087] This embodiment uses an integrated model characterized by the actual physical parameters of a mobile composite robot and whole-body control as the basic control framework. After receiving robot operation task data, it generates the robot's total control quantity based on the whole-body control framework, and then performs inverse analytical calculations to obtain the control quantities of the robot's moving body and the robotic arm's operation control quantities. These control quantities are sent to the wheel hub drivers of the moving body and the joint drivers of the robotic arm to execute corresponding actions. In this calculation of the total control quantity, the calculation of the robot's moving body and robotic arm control quantities is not distinguished separately; only after the total control quantity is finally obtained, the moving body control quantity and the robotic arm control quantity are analyzed based on this total control quantity. This approach is fundamentally different from traditional mobile composite robot control. In this control mode, the robot's walking and operation actions have strong synchronization and coordination. For example, when the mobile composite robot approaches the operation target, it adjusts the pose of the moving body and the robotic arm simultaneously, so that the robotic arm has the optimal operation posture the moment it reaches the operation target. This improves the robot's work efficiency and avoids the drawback of discovering that the robotic arm is in a singularity zone and unable to operate the target only when it reaches the operation target. Traditionally controlled mobile composite robots typically stop moving upon reaching the target, then adjust their orientation; once the orientation is corrected, the robot remains stationary while the robotic arm begins its preparation for the target. This mode represents typical sequential control and is inefficient.

[0088] In this embodiment, after receiving an external task command, the robot controller parses it into input control quantities for the whole-body controller. Simultaneously, these control quantities serve as input to the integrated system model. Robot state parameters, including current walking speed, angle, acceleration, and robotic arm motion parameters, are synchronously input into the integrated system model. Based on the input control quantities and the system theory output from the integrated system model, the whole-body controller adjusts the vector parameters and the robot's perception data to operate. The output of the whole-body controller serves as the control quantities for the robot's walking and robotic arm operations. A dynamic weight adjustment function F(w) is included in the whole-body controller to optimize the overall control effect.

[0089] In this embodiment, the process for optimizing the whole-body control effect is as follows: Figure 4 As shown, the whole-body controller evaluates the operation results based on the historical index parameter data of the mobile composite robot, and determines whether to optimize the index parameter data corresponding to the operation results based on the evaluation results, and adds the optimization weight parameters to the generation of the total control quantity.

[0090] In this embodiment, the historical performance parameters mainly include parameters that characterize the robot's operational performance, such as the robot's cycle time, operation duration, operation error, and end-effector stability. This embodiment evaluates the historical operation results of the mobile composite robot based on these historical parameters, following certain rules. Each performance parameter is evaluated using three levels: Excellent, Good, and Average. Of course, other evaluation levels or rules can be set according to actual needs. A performance parameter reaching Excellent means no further optimization is needed, while a Good or Average performance parameter requires optimization and improvement. Each performance parameter is associated with a corresponding task (Task X) or several tasks and their associated weight parameters. Adjusting the corresponding weight parameter values ​​optimizes the corresponding performance parameter. The adjustment range of the weight corresponding to "Good" is generally greater than that corresponding to "Average," but to ensure overall performance stability, the adjustment range will not be too large. Performance improvement is ultimately achieved through continuous iterative calculation. Therefore, under this mechanism, the weight parameters are always in dynamic adjustment, thus playing a role in dynamically optimizing performance.

[0091] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for whole-body control of a mobile composite robot, characterized in that, The steps include the following: S1, Establish an integrated system model of the mobile composite robot in the world coordinate system; S2, after receiving external data instructions, the robot controller parses the external data instructions into input control quantities for the whole-body controller and transmits them to the whole-body controller and the integrated system model; S3, the mobile composite robot transmits its state parameters to the integrated system model; the mobile composite robot transmits its perception data to the whole-body controller; S4, the integrated system model transmits the generated system theoretical adjustment vector parameters to the whole-body controller; S5, the whole-body controller generates the total control quantity based on the input control quantity, the system theory, the adjusted vector parameters, and the robot's perception data; S6, the whole-body controller performs inverse analytical calculations to obtain the control quantities of the mobile body of the mobile composite robot and the operation control quantities of the robotic arm, respectively; S7, the whole body controller transmits the control quantities of the mobile body of the mobile composite robot and the operation control quantities of the robotic arm to the hub driver corresponding to the mobile body and the servo joint driver corresponding to the robotic arm, respectively. S5 includes a dynamic weight optimization strategy; The whole-body controller evaluates the operation results based on the historical index parameter data of the mobile composite robot, and determines whether to optimize the index parameter data corresponding to the operation results based on the evaluation results, and adds the dynamically optimized weight parameters to the generation of the total control quantity.

2. The whole-body control method for a mobile composite robot according to claim 1, characterized in that, In step S5, the whole-body controller establishes a multi-task priority null space projection method based on the input control quantity, system theoretical adjustment vector parameters, and sensing data. The low-priority tasks are placed in the null space of the high-priority tasks for calculation, and then the control quantities of all tasks are added together to obtain the total control quantity.

3. The whole-body control method for a mobile composite robot according to claim 2, characterized in that, In step S1, the integrated system model is coupled with the mathematical model of the mobile body of the mobile composite robot and the mathematical model of the robotic arm. The mathematical model of the mobile body of the mobile composite robot is as follows: Where A and B are parameter matrices related to the mechanical structure of the robot body; The vector of rotational speed of the robot's left and right drive wheels; The angular velocity of the moving body. Let be the differential of the position parameters of the robotic arm base on the moving body; the mathematical model of the robotic arm is: in, This is the pose vector of the robotic arm's end effector; A function representing the known structural parameters of the robotic arm, used to describe the pose behavior of the robotic arm's end effector; the integrated system model of the mobile composite robot is: in, This represents the velocity vector matrix at the end effector of the mobile composite robot arm. θ represents the Jacobian matrix that simultaneously couples the mobile body's pose and the robot arm's end effector pose angle; k is the mobile composite robot parameter matrix; θ is the robot arm pose angle. This represents the velocity vector of the robotic arm's motion pose.

4. The whole-body control method for a mobile composite robot according to claim 3, characterized in that, The mathematical expression for the whole-body controller is: Where τ is the planning torque weighted by the weights; ω1, ω2, ..., ω k Dynamically optimized weight coefficients for tasks corresponding to different historical indicator parameter data; J1, J2, ..., J k Jacobian matrices for tasks corresponding to different historical indicator parameters; F1, F2, ..., F k This represents the Cartesian impedance space matrix corresponding to different historical index parameter data.

5. The whole-body control method for a mobile composite robot according to claim 4, characterized in that, The state parameters of the mobile composite robot include the current walking speed, angle, acceleration, and robotic arm motion parameters.

6. The whole-body control method for a mobile composite robot according to claim 5, characterized in that, The historical performance parameters of the mobile composite robot include operation cycle time, operation duration, operation error, and stability of the robotic arm end effector.

7. The whole-body control method for a mobile composite robot according to claim 6, characterized in that, The evaluation results are categorized into three levels: Excellent, Good, and Average. The indicator parameters corresponding to the operation results with an Excellent level will not be optimized further. The indicator parameters corresponding to the operation results with a Good level will undergo secondary optimization. The indicator parameters corresponding to the operation results with an Average level will undergo primary optimization.

Images