Multi-mechanical-arm synchronous control system based on position instruction

Abstract

The invention relates to the technical field of automatic control, and discloses a multi-mechanical-arm synchronous control system based on a position instruction, which comprises a track discretization module, a synchronous clock signal generation module, an evaluation error module, an error compensation module, an AI error prediction module and a track optimization module, and by introducing a feedback error compensation mechanism and an AI prediction model, the position errors of the multiple mechanical arms in the synchronization process can be greatly reduced. The adaptability is high, the system can adapt to mechanical arms with different speed characteristics, and personalized trajectory planning schemes are designed for different mechanical arms. A speed control interface is not needed, and the system only depends on a position information instruction for control and is suitable for a mechanical arm system which cannot be controlled through speed.

Description

Technical Field

[0001] The present invention relates to the field of automation control technology, and in particular to a multi-robot arm synchronous control system based on position instructions. Background Art

[0002] With the rapid development of industrial automation and robotics, multi-robot collaborative operations have been widely used in manufacturing, logistics, assembly and processing, etc. Existing multi-robot control systems usually rely on precise speed, acceleration and position control to ensure that the multi-robots can keep pace with each other on the trajectory.

[0003] However, in some application scenarios, the control system of the robot arm only supports instructions through position information, but cannot directly control the speed of the robot arm. Since the speed of the robot arm is uncontrollable, the traditional multi-robot arm synchronization method is difficult to apply, making it difficult to achieve synchronous collaboration of multiple robot arms, affecting production efficiency and control accuracy.

[0004] Therefore, how to achieve synchronous control of multiple robotic arms through position information instructions when the speed is uncontrollable is a technical problem that needs to be solved urgently. Summary of the invention

[0005] The technical problem to be solved by the present invention is: how to solve the existing problem of realizing synchronous control of multiple robotic arms through position information instructions when the speed is uncontrollable, and provides a multi-robotic arm synchronous control system based on position instructions.

[0006] The present invention solves the above technical problems through the following technical solutions: a multi-manipulator synchronous control system based on position instructions, including a trajectory discretization module, a synchronous clock signal generation module, an evaluation error module, an error compensation module, an AI error prediction module and a trajectory optimization module;

[0007] The trajectory discretization module is used to discretize the target trajectory of each robot arm into multiple key points, and use the multiple key points as the target position of each robot arm in each clock cycle;

[0008] The synchronous clock signal generating module is used to generate a global synchronous clock signal so that all the robot arms receive new target position instructions within the same clock cycle;

[0009] The evaluation error module compares the actual position of each robot arm with the preset target position at the end of each clock cycle to obtain an evaluation error;

[0010] The error compensation module adjusts the target position of each robotic arm at the next clock cycle according to the evaluation error of each robotic arm at the end of each clock cycle to implement error compensation;

[0011] The AI ​​error prediction module predicts the evaluation error of the next clock cycle and dynamically adjusts the target position based on the historical operation data and current state of each robotic arm;

[0012] The trajectory optimization module is used to optimize the trajectory synchronization strategy of multiple robotic arms.

[0013] Optionally, the control system further comprises a real-time feedback module, which is used to obtain the actual position of each robotic arm and transmit the actual position to the error evaluation module.

[0014] Optionally, the specific process of predicting the evaluation error of the next clock cycle and dynamically adjusting the target position is as follows:

[0015] Based on historical operation data and current status, the error prediction algorithm is trained through supervised learning or deep learning model, and the error prediction algorithm is used to predict the error trend of each robot arm in the next clock cycle;

[0016] According to the predicted error trend, the target position in the next clock cycle is adjusted in advance.

[0017] Optionally, the historical operation data includes the actual position, target position, evaluation error and error compensation of each robot arm in the past multiple preset clock cycles;

[0018] The current state contains the actual position, target position, and estimated error of each robot in the current clock cycle.

[0019] Optionally, the optimization method of the trajectory synchronization strategy includes trajectory optimization of reinforcement learning and nonlinear trajectory planning.

[0020] Optionally, the reinforcement learning trajectory optimization includes optimizing the target trajectories of the multiple robotic arms using a preset reinforcement learning model so as to synchronize the target trajectories of the multiple robotic arms.

[0021] Optionally, the nonlinear trajectory planning includes dynamic trajectory adjustment and collaborative trajectory generation;

[0022] The dynamic trajectory adjustment process is to dynamically generate corresponding nonlinear trajectories according to the different speed characteristics of the robot arm.

[0023] The collaborative trajectory generation process is based on the AI ​​error prediction module. According to the actual operation trajectories of multiple robotic arms, multiple robotic arms are dynamically adjusted in each clock cycle to generate a preset trajectory adjustment plan.

[0024] Compared with the prior art, the present invention has the following advantages:

[0025] The synchronization accuracy is improved. By introducing feedback error compensation mechanism and AI prediction model, the position error of multiple robotic arms during the synchronization process can be greatly reduced.

[0026] The system has strong adaptability and can adapt to robotic arms with different speed characteristics and design personalized trajectory planning solutions for different robotic arms.

[0027] No speed control interface is required, the system only relies on position information instructions for control, and is suitable for robotic arm systems that cannot be controlled by speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is the overall flow chart of the present invention. DETAILED DESCRIPTION

[0029] The following is a detailed description of an embodiment of the present invention. This embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation method and a specific operation process are given, but the protection scope of the present invention is not limited to the following embodiment.

[0030] like Figure 1 As shown, this embodiment provides a technical solution: a multi-manipulator synchronous control system based on position instructions, including a trajectory discretization module, a synchronous clock signal generation module, an evaluation error module, an error compensation module, an AI error prediction module and a trajectory optimization module;

[0031] The trajectory discretization module is used to discretize the target trajectory of each robot arm into multiple key points, and use the multiple key points as the target position of each robot arm in each clock cycle;

[0032] The synchronous clock signal generating module is used to generate a global synchronous clock signal so that all the robot arms receive new target position instructions within the same clock cycle;

[0033] The evaluation error module compares the actual position of each robotic arm with the preset target position at the end of each clock cycle to obtain an evaluation error; the control system also includes a real-time feedback module, which is used to obtain the actual position of each robotic arm and transmit the actual position to the evaluation error module.

[0034] The real-time feedback module can be a sensor on the robotic arm (such as an encoder, a position sensor, etc.), which can accurately monitor and obtain the actual position of the robotic arm in real time.

[0035] The error compensation module adjusts the target position of each robot arm in the next clock cycle according to the evaluation error of each robot arm at the end of each clock cycle to implement error compensation; if a robot arm fails to reach the preset target position on time after a clock cycle, the target position issued by it will be adjusted in the next cycle, and the amount of position adjustment is the error compensation, so that it can gradually restore synchronization. The adjustment of the target position can be carried out step by step, and the error in one clock cycle can be compensated in multiple subsequent clock cycles, so that the actual position and the target position can be synchronized in the end.

[0036] The AI ​​error prediction module predicts the evaluation error of the next clock cycle and dynamically adjusts the target position based on the historical operation data and current state of each robotic arm; the historical operation data includes the actual position, target position, evaluation error and error compensation of each robotic arm in the past multiple preset clock cycles;

[0037] The current state contains the actual position, target position, and estimated error of each robot in the current clock cycle.

[0038] The AI ​​error prediction module can provide model data for the AI ​​error prediction model. The historical operation data and current status provide model data for the AI ​​error prediction model, thereby making the model prediction more accurate.

[0039] The specific process of the AI ​​error prediction model predicting the evaluation error of the next clock cycle and dynamically adjusting the target position is as follows:

[0040] Based on historical operation data and current status, the error prediction algorithm is trained through supervised learning or deep learning model, and the error prediction algorithm is used to predict the error trend of each robotic arm in the next clock cycle; specifically, deep neural network is used for error prediction.

[0041] According to the predicted error trend, the target position in the next clock cycle is adjusted in advance.

[0042] The trajectory optimization module is used to optimize the trajectory synchronization strategy of multiple robotic arms. Specifically, the optimization method of the trajectory synchronization strategy includes trajectory optimization of reinforcement learning and nonlinear trajectory planning.

[0043] The reinforcement learning trajectory optimization includes optimizing the target trajectories of multiple robotic arms using a preset reinforcement learning model to synchronize the target trajectories of the multiple robotic arms, and the reinforcement learning model is used based on a reinforcement learning algorithm.

[0044] Specifically, the nonlinear trajectory planning includes dynamic trajectory adjustment and collaborative trajectory generation;

[0045] The dynamic trajectory adjustment process dynamically generates corresponding nonlinear trajectories for robotic arms with different speed characteristics. Specifically, a nonlinear trajectory planning algorithm is used to generate personalized nonlinear trajectories for different robotic arms. These nonlinear trajectories can be different in the middle section according to the current state of the robotic arm, but they are synchronized at the key position points (target positions).

[0046] The collaborative trajectory generation process is based on the AI ​​error prediction module. According to the actual operation trajectories of multiple robotic arms, multiple robotic arms are dynamically adjusted in each clock cycle to generate a preset trajectory adjustment plan.

[0047] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0048] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0049] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A multi-manipulator synchronous control system based on position instructions, characterized in that: It includes trajectory discretization module, synchronous clock signal generation module, error evaluation module, error compensation module, AI error prediction module and trajectory optimization module; The trajectory discretization module is used to discretize the target trajectory of each robot arm into multiple key points, and use the multiple key points as the target position of each robot arm in each clock cycle; The synchronous clock signal generating module is used to generate a global synchronous clock signal so that all the robot arms receive new target position instructions within the same clock cycle; The evaluation error module compares the actual position of each robot arm with the preset target position at the end of each clock cycle to obtain an evaluation error; The error compensation module adjusts the target position of each robotic arm at the next clock cycle according to the evaluation error of each robotic arm at the end of each clock cycle to implement error compensation; The AI ​​error prediction module predicts the evaluation error of the next clock cycle and dynamically adjusts the target position based on the historical operation data and current state of each robotic arm; The trajectory optimization module is used to optimize the trajectory synchronization strategy of multiple robotic arms.

2. A multi-manipulator synchronous control system based on position instructions according to claim 1, characterized in that: The control system further comprises a real-time feedback module, which is used to obtain the actual position of each robot arm and transmit the actual position to the error evaluation module.

3. The multi-manipulator synchronous control system based on position instructions according to claim 1, characterized in that: The specific process of predicting the evaluation error of the next clock cycle and dynamically adjusting the target position is as follows: Based on historical operation data and current status, the error prediction algorithm is trained through supervised learning or deep learning model, and the error prediction algorithm is used to predict the error trend of each robot arm in the next clock cycle; According to the predicted error trend, the target position in the next clock cycle is adjusted in advance.

4. The multi-manipulator synchronous control system based on position instructions according to claim 1, characterized in that: The historical operation data includes the actual position, target position, evaluation error and error compensation of each robot arm in the past multiple preset clock cycles; The current state contains the actual position, target position, and estimated error of each robot in the current clock cycle.

5. The multi-manipulator synchronous control system based on position instructions according to claim 1, characterized in that: The optimization method of the trajectory synchronization strategy includes trajectory optimization of reinforcement learning and nonlinear trajectory planning.

6. The multi-manipulator synchronous control system based on position instructions according to claim 5, characterized in that: The reinforcement learning trajectory optimization includes optimizing the target trajectories of multiple robotic arms using a preset reinforcement learning model so as to synchronize the target trajectories of the multiple robotic arms.

7. The multi-manipulator synchronous control system based on position instructions according to claim 5, characterized in that: The nonlinear trajectory planning includes dynamic trajectory adjustment and collaborative trajectory generation; The dynamic trajectory adjustment process is to dynamically generate corresponding nonlinear trajectories according to the different speed characteristics of the robot arm; The collaborative trajectory generation process is based on the AI ​​error prediction module. According to the actual operation trajectories of multiple robotic arms, the trajectories of multiple robotic arms are dynamically adjusted in each clock cycle to generate a preset trajectory adjustment plan.

Images