Inspection robot fault diagnosis method and system based on multi-sensor fusion

Abstract

The invention relates to the technical field of robots and fault diagnosis, and the method comprises the following steps: collecting motion data and external environment information of an inspection robot; establishing a prediction model of the inspection robot, and generating a prediction control signal according to the prediction model; according to the error between the prediction control signal and the actual control signal, control parameters of the inspection robot are adjusted through a self-adaptive algorithm; generating a cooperative control signal of the inspection robot according to the adjusted control parameters; and sending the cooperative control signal to an execution mechanism of the inspection robot to realize cooperative movement of the inspection robot. The method can effectively overcome the defects of the traditional fault diagnosis method in the aspects of model uncertainty, external interference and the like, thereby improving the control precision and efficiency of the inspection robot. In addition, the multi-sensor fusion fault diagnosis method can also consider the synergistic effect between the inspection robots, so that the movement of the inspection robots is more coordinated, and the performance of the robots is improved.

Description

Technical Field

[0001] The present invention relates to the field of robot technology and fault diagnosis technology, and in particular to a patrol robot fault diagnosis method and system based on multi-sensor fusion. Background Art

[0002] With the rapid development of robot technology, inspection robots are increasingly used in industrial production, service industry and other fields. Dual-arm robots are a new type of inspection robot that is widely used. Among the fault diagnosis methods of inspection robots, collaborative fault diagnosis methods have always been the focus and difficulty of research. Traditional collaborative fault diagnosis methods for inspection robots mainly rely on accurate models and algorithms. However, in practical applications, due to the influence of various factors, such as load changes of robots and external interference, traditional fault diagnosis methods are difficult to meet the requirements of high precision and high efficiency, and are prone to failure.

[0003] The traditional collaborative fault diagnosis method of inspection robots has the following shortcomings: First, traditional fault diagnosis methods often require precise models and algorithms. However, in practical applications, due to the influence of various factors, such as robot load changes, external interference, etc., traditional fault diagnosis methods are difficult to meet the requirements of high precision and high efficiency; secondly, traditional fault diagnosis methods often ignore the synergy between inspection robots, resulting in uncoordinated movement of inspection robots, thus affecting the performance of the robots; thirdly, traditional fault diagnosis methods often require a lot of calculations, resulting in poor real-time performance, which is difficult to meet the needs of practical applications.

[0004] In order to solve the above problems, multi-sensor fusion fault diagnosis methods have gradually become a research hotspot in recent years. By introducing prediction models and adaptive algorithms, multi-sensor fusion fault diagnosis methods can effectively overcome the shortcomings of traditional fault diagnosis methods in terms of model uncertainty and external interference, thereby improving the control accuracy and efficiency of inspection robots. In addition, multi-sensor fusion fault diagnosis methods can also consider the synergy between inspection robots, making the movement of inspection robots more coordinated, thereby improving the performance of robots.

[0005] However, the existing multi-sensor fusion fault diagnosis method still has some problems, such as high computational complexity, poor real-time performance, etc. In view of these problems, the present invention proposes a patrol robot fault diagnosis method and system based on multi-sensor fusion, which can reduce computational complexity and improve real-time performance while ensuring control accuracy and efficiency. Summary of the invention

[0006] In view of the above existing problems, the present invention is proposed.

[0007] Therefore, the present invention provides a patrol robot fault diagnosis method based on multi-sensor fusion. Through multi-sensor fusion technology, the accuracy, flexibility and stability of the collaborative control of the patrol robot are improved, the problem of collaborative operation in a changing environment is solved, and the operation efficiency and safety are improved.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solutions: a patrol robot fault diagnosis method based on multi-sensor fusion, comprising: collecting motion data and external environment information of the patrol robot; establishing a prediction model of the patrol robot, and generating a prediction control signal according to the prediction model; adjusting the control parameters of the patrol robot through an adaptive algorithm according to the error between the prediction control signal and the actual control signal; generating a collaborative control signal of the patrol robot according to the adjusted control parameters; and sending the collaborative control signal to the actuator of the patrol robot to realize the collaborative movement of the patrol robot.

[0009] As a preferred solution of the inspection robot fault diagnosis method based on multi-sensor fusion described in the present invention, the collection of motion data and external environment information includes: using sensors and monitoring equipment to collect the motion data of the inspection robot, the motion data includes position, speed and acceleration, using force sensors to measure the torque and force at the joints, and the force of the end effector; at the same time, collecting external environment information, capturing the three-dimensional structure of the surrounding environment, obtaining the pressure and shape information of the contact with the object, and monitoring the ambient temperature and humidity;

[0010] Appropriate sensors are installed on the joints and end effectors of the inspection robot, and these sensors are connected to the control unit; the sensor data are transmitted to the control unit in real time via wireless means, and the collected data is processed in the control unit.

[0011] As a preferred solution of the inspection robot fault diagnosis method based on multi-sensor fusion described in the present invention, wherein: the establishment of the prediction model of the inspection robot includes selecting initial parameters and collecting data from the movement of the inspection robot;

[0012] The data includes motion trajectory, force, torque, and external environment information. The objective function J(θ) is constructed based on the collected data:

[0013]

[0014] g k (θ)≤0,k=1,2,…,K

[0015] Among them, y i represents the actual observed output value, represents the predicted output value, λ regularization parameter, μ j(θ,t) is the dynamic characteristic function, Δt represents the prediction time range, M represents the number of dynamic characteristic functions, N represents the number of data points, and g k (θ) represents the inequality constraint, k represents the number of inequality constraints;

[0016] Ensure that the robot does not violate the laws of physics during movement, the force does not exceed the maximum bearing capacity of the actuator, and the joint angle does not exceed the range of motion. Set dynamic constraints:

[0017] g1(θ)=min(F max -F(θ),0)≤0

[0018] g2(θ)=min(τ max -τ(θ),0)≤0

[0019] g3(θ)=min(γ max -γ(θ),0)≤0

[0020] Where F(θ) is the force calculated from the model parameters θ, and F max is the maximum allowable force, τ(θ) is the moment calculated from the model parameter θ, τ max is the maximum allowable moment, γ(θ) is the joint angle calculated from the model parameter θ, and γ max is the maximum allowed joint angle;

[0021] To ensure that the robot's motion trajectory does not collide with itself or enter areas outside the workspace, set kinematic constraints:

[0022] g4(θ)=min(d self (θ),0)≤0

[0023] g5(θ)=min(d ws (θ),0)≤0

[0024] Among them, d self (θ) represents self-collision detection, d ws (θ) represents the workspace boundary detection;

[0025] Ensure that the robot's performance indicators when performing tasks meet specific requirements and set performance constraints:

[0026] g6(θ)=T(θ)-T max ≤0

[0027] g7(θ)=E(θ)-E max ≤0

[0028] Among them, T(θ) represents the time required to complete the task, T maxrepresents the maximum allowed time, E(θ) represents the energy consumption required to complete the task, and E max Indicates the maximum allowable energy consumption;

[0029] To ensure smooth robot motion and avoid sudden acceleration or deceleration, set smoothness performance constraints:

[0030]

[0031] Among them, σ 2 Represents the upper limit of the smoothness indicator;

[0032] To ensure that the two arms of the inspection robot maintain coordination during movement and avoid conflicts or uncoordinated movements, set coordination constraints:

[0033]

[0034] Among them, y left and right Represent the predicted outputs of the left and right arms, δ 2 Indicates the upper limit of the synergy index.

[0035] As a preferred solution of the inspection robot fault diagnosis method based on multi-sensor fusion described in the present invention, wherein: generating the prediction control signal includes predicting the state of the inspection robot in the future period of time through a prediction model;

[0036] According to the predicted future state, the control signal required to achieve the desired target state is calculated. The control signal includes the joint torque, joint angle target and the pose target of the end effector:

[0037] u pred (t) = C(x pred (t),x des (t))

[0038] Among them, u pred (t) represents the predicted control signal generated at time t, C is the control strategy function, which calculates the control signal based on the predicted state and the desired state, x pred (t) represents the predicted robot state at time t, x des (t) represents the desired robot state at time t;

[0039] The predicted control signal is sent to the robot's control system. At the same time, the control system generates an actual control signal based on the current sensor data. The actual control signal drives the motor and actuator to make the robot's actual state reach the predicted state:

[0040] u act (t) = K(t)·(x des (t)-xact (t))+u pred (t)

[0041] K(t)=P(t)C T (x des (t))C(x(t))

[0042] Among them, u act (t) represents the control signal actually applied at time t, x act (t) represents the state of the actual application at time t, K(t) represents the feedback gain matrix, P(t) represents the prediction error covariance matrix, C(x des (t)) represents the transition matrix of the desired state, and C(x(t)) represents the transition matrix of the actual state.

[0043] As a preferred solution of the inspection robot fault diagnosis method based on multi-sensor fusion described in the present invention, wherein: the adaptive algorithm adjusts the control parameters including calculating the prediction error;

[0044] e(t)=x(t)-x pred (t)

[0045] Among them, x(t) represents the actual state;

[0046] A dynamically changing prediction error threshold Δe is set to determine whether the prediction error is within an acceptable range. When the prediction error e(t) exceeds the dynamic threshold Δe, the prediction error processing mechanism is started, and a dynamically changing prediction error covariance matrix threshold is set to determine whether the prediction error covariance matrix is ​​stable.

[0047] Compute the forecast error covariance matrix:

[0048] P(t)=Cov(x(t),x pred (t))

[0049] A dynamically changing prediction error covariance matrix threshold ΔP is set to determine whether the prediction error covariance matrix is ​​stable. When the change rate of the prediction error covariance matrix P(t) exceeds the dynamic threshold ΔP, the prediction error covariance matrix processing mechanism is started. According to the results of the prediction error processing module and the prediction error covariance matrix processing module, the control parameters are adjusted. The amplitude and frequency of adjusting the control parameters are proportional to the processing results of the prediction error and the prediction error covariance matrix.

[0050] As a preferred solution of the inspection robot fault diagnosis method based on multi-sensor fusion described in the present invention, wherein: the generation of the coordinated control signal of the inspection robot includes using distributed computing technology to distribute the control task to multiple computing units and coordinating the global control strategy through network communication;

[0051] Decompose the control task into multiple subtasks. Each subtask is handled by a computing unit. The subtasks are reasonably allocated according to the computing power of the computing unit and the network delay. Each computing unit generates a local control signal based on the subtask and local sensor data.

[0052] Through network communication, all computing units are ensured to have the same time reference and status information. The computing units communicate through the network to exchange control signals and coordinate the global control strategy. Each computing unit fuses the generated local control signal with the control signals from other computing units to generate a global control signal:

[0053]

[0054] Among them, u glob Represents the global control signal, u loc represents the local control signal, α i is the weight of the ith computing unit, t i is the sampling time of the ith computing unit;

[0055] The global control signal is sent to the actuator of the inspection robot, which performs corresponding movements according to the control signal to realize the coordinated movement of the inspection robot; the actual motion data of the inspection robot is collected through sensors, the actual motion data is compared with the expected motion data, the control error is calculated, the control strategy is adjusted according to the control error, and a new coordinated control signal is generated.

[0056] As a preferred solution of the inspection robot fault diagnosis method based on multi-sensor fusion described in the present invention, wherein: the execution collaborative control signal includes: the actuator receives the control signal from the computing unit, and the control unit in the actuator is responsible for analyzing the signal and converting it into a drive signal of the motor and a control signal of the actuator;

[0057] The control unit in the actuator adjusts the movement of the motor and actuator according to the analyzed control signal. The motor control unit converts the signal into the current or speed control signal of the motor to accurately control the movement of each joint; the actuator control unit converts the signal into the force or displacement control signal of the actuator to accurately control the operation of the end effector.

[0058] The sensors on the inspection robot monitor the robot's motion status and external environment information in real time. The sensor data is fed back to the control unit of the actuator through the data acquisition system. The control unit evaluates the effect of the control signal based on the feedback data and uses it to adjust future control strategies.

[0059] Another object of the present invention is to provide a patrol robot collaborative control system based on multi-sensor fusion, which can reduce computational complexity and improve real-time performance while ensuring control accuracy and efficiency.

[0060] As a preferred solution of the inspection robot collaborative control system based on multi-sensor fusion described in the present invention, it includes: a data acquisition module, a prediction model module, a control strategy module, an actuator module, a communication module, and an adaptive algorithm module;

[0061] The data acquisition module collects the status data and external environment data of the inspection robot in real time;

[0062] The prediction model module establishes a prediction model, which predicts the future state of the inspection robot based on the current state and external environment information;

[0063] The control strategy module defines the control strategy, adjusts the control parameters according to the prediction error and the covariance matrix of the prediction error, and generates the collaborative control signal of the inspection robot;

[0064] The actuator module receives the coordinated control signal, adjusts the joint angle of the inspection robot, the position and torque of the end effector, and realizes the coordinated movement of the inspection robot;

[0065] The communication module implements network communication between computing units, ensures data synchronization and transmission of control signals, and coordinates global control strategies;

[0066] The adaptive algorithm module adjusts the control parameters to realize adaptive control according to the prediction error and the covariance matrix of the prediction error.

[0067] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, it implements the steps of a fault diagnosis method for an inspection robot based on multi-sensor fusion.

[0068] A computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the computer program implements the steps of a fault diagnosis method for an inspection robot based on multi-sensor fusion.

[0069] The beneficial effects of the present invention are as follows: by introducing a prediction model and an adaptive algorithm, the present method can effectively overcome the deficiencies of traditional fault diagnosis methods in terms of model uncertainty and external interference, thereby improving the control accuracy and efficiency of the inspection robot; by taking into account the synergy between the inspection robots, the movement of the inspection robots is more coordinated, thereby improving the performance of the robots; by optimizing the adaptive algorithm and the adjustment strategy of the control parameters, the computational complexity is reduced and the real-time performance is improved; the present invention is not only applicable to inspection robots, but can also be extended to multi-arm robots or other types of robot systems, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0071] Figure 1 A schematic flow chart of a fault diagnosis method for an inspection robot based on multi-sensor fusion is provided as an embodiment of the present invention.

[0072] Figure 2 A time series diagram of a fault diagnosis method for an inspection robot based on multi-sensor fusion provided by an embodiment of the present invention.

[0073] Figure 3 An assembly error diagram of a fault diagnosis method for an inspection robot based on multi-sensor fusion provided by an embodiment of the present invention.

[0074] Figure 4 A schematic diagram of working modules of a patrol robot collaborative control system based on multi-sensor fusion provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0075] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0076] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0077] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0078] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0079] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0080] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0081] Example 1

[0082] Reference Figure 1 , which is the first embodiment of the present invention, and provides a fault diagnosis method for an inspection robot based on multi-sensor fusion, comprising:

[0083] S1: Collect the motion data and external environment information of the inspection robot.

[0084] Furthermore, the collection of motion data and external environment information includes using sensors and monitoring equipment to collect motion data of the inspection robot, including position, speed and acceleration, using force sensors to measure the torque and force at the joints, and the force of the end effector; at the same time, collecting external environment information, capturing the three-dimensional structure of the surrounding environment, obtaining pressure and shape information of contact with objects, and monitoring the ambient temperature and humidity;

[0085] Appropriate sensors are installed on the joints and end effectors of the inspection robot, and these sensors are connected to the control unit; the sensor data is transmitted to the control unit in real time via wireless means, and the collected data is processed in the control unit to improve the data quality.

[0086] S2: Establish a prediction model for the inspection robot and generate a prediction control signal based on the prediction model.

[0087] Furthermore, the establishment of the prediction model of the inspection robot includes selecting initial parameters, collecting data from the movement of the inspection robot, including movement trajectory, force, torque, and external environment information, and constructing the objective function J(θ) based on the collected data:

[0088]

[0089] g k (θ)≤0,k=1,2,…,K

[0090] Among them, y i represents the actual observed output value, represents the predicted output value, λ regularization parameter, μ j (θ,t) is the dynamic characteristic function, Δt represents the prediction time range, M represents the number of dynamic characteristic functions, N represents the number of data points, and g k (θ) represents the inequality constraint, k represents the number of inequality constraints;

[0091] Ensure that the robot does not violate the laws of physics during movement, the force does not exceed the maximum bearing capacity of the actuator, and the joint angle does not exceed the range of motion. Set dynamic constraints:

[0092] g1(θ)=min(F max -F(θ),0)≤0

[0093] g2(θ)=min(τ max -τ(θ),0)≤0

[0094] g3(θ)=min(γ max -γ(θ),0)≤0

[0095] Where F(θ) is the force calculated from the model parameters θ, and F max is the maximum allowable force, τ(θ) is the moment calculated from the model parameter θ, τ max is the maximum allowable moment, γ(θ) is the joint angle calculated from the model parameter θ, and γ max is the maximum allowed joint angle;

[0096] To ensure that the robot's motion trajectory does not collide with itself or enter areas outside the workspace, set kinematic constraints:

[0097] g4(θ)=min(d self (θ),0)≤0

[0098] g5(θ)=min(d ws (θ),0)≤0

[0099] Among them, d self (θ) represents self-collision detection, d ws (θ) represents the workspace boundary detection;

[0100] Ensure the robot's performance indicators when performing tasks, meet specific requirements, and set performance constraints;

[0101] g6(θ)=T(θ)-T max ≤0

[0102] g7(θ)=E(θ)-E max ≤0

[0103] Among them, T(θ) represents the time required to complete the task, T max represents the maximum allowed time, E(θ) represents the energy consumption required to complete the task, and E max Indicates the maximum allowable energy consumption;

[0104] To ensure smooth robot motion and avoid sudden acceleration or deceleration, set smoothness performance constraints:

[0105]

[0106] Among them, σ 2 Represents the upper limit of the smoothness indicator;

[0107] To ensure that the two arms of the inspection robot maintain coordination during movement and avoid conflicts or uncoordinated movements, set coordination constraints:

[0108]

[0109] Among them, y left and right Represent the predicted outputs of the left and right arms, δ 2Indicates the upper limit of the synergy index.

[0110] Furthermore, the generating of the prediction control signal includes predicting the state of the inspection robot in a future period of time through a prediction model;

[0111] According to the predicted future state, the control signal required to achieve the desired target state is calculated. The control signal includes the joint torque, joint angle target and the pose target of the end effector:

[0112] u pred (t) = C(x pred (t),x des (t))

[0113] Among them, u pred (t) represents the predicted control signal generated at time t, C is the control strategy function, which calculates the control signal based on the predicted state and the desired state, x pred (t) represents the predicted robot state at time t, x des (t) represents the desired robot state at time t;

[0114] The predicted control signal is sent to the robot's control system. At the same time, the control system generates an actual control signal based on the current sensor data. The actual control signal drives the motor and actuator to make the robot's actual state reach the predicted state:

[0115] u act (t) = K(t)·(x des (t)-x act (t))+u pred (t)

[0116] K(t)=P(t)C T (x des (t))C(x(t))

[0117] Among them, u act (t) represents the control signal actually applied at time t, x act (t) represents the state of the actual application at time t, K(t) represents the feedback gain matrix, P(t) represents the prediction error covariance matrix, C(x des (t)) represents the transition matrix of the desired state, and C(x(t)) represents the transition matrix of the actual state.

[0118] S3: According to the error between the predicted control signal and the actual control signal, the control parameters of the inspection robot are adjusted through an adaptive algorithm.

[0119] Furthermore, the adaptive algorithm adjusts the control parameters including calculating the prediction error:

[0120] e(t)=x(t)-x pred (t)

[0121] Among them, x(t) represents the actual state;

[0122] A dynamically changing prediction error threshold Δe is set to determine whether the prediction error is within an acceptable range. When the prediction error e(t) exceeds the dynamic threshold Δe, the prediction error processing mechanism is started, and a dynamically changing prediction error covariance matrix threshold is set to determine whether the prediction error covariance matrix is ​​stable.

[0123] Compute the forecast error covariance matrix:

[0124] P(t)=Cov(x(t),x pred (t))

[0125] A dynamically changing prediction error covariance matrix threshold ΔP is set to determine whether the prediction error covariance matrix is ​​stable. When the change rate of the prediction error covariance matrix P(t) exceeds the dynamic threshold ΔP, the prediction error covariance matrix processing mechanism is started. According to the results of the prediction error processing module and the prediction error covariance matrix processing module, the control parameters are adjusted. The amplitude and frequency of adjusting the control parameters are proportional to the processing results of the prediction error and the prediction error covariance matrix.

[0126] S4: Generate a collaborative control signal for the inspection robot according to the adjusted control parameters.

[0127] Furthermore, the generating of the coordinated control signal of the inspection robot includes distributing the control task to multiple computing units using distributed computing technology and coordinating the global control strategy through network communication;

[0128] Decompose the control task into multiple subtasks. Each subtask is handled by a computing unit. The subtasks are reasonably allocated according to the computing power of the computing unit and the network delay. Each computing unit generates a local control signal based on the subtask and local sensor data.

[0129] Through network communication, all computing units are ensured to have the same time reference and status information. The computing units communicate through the network to exchange control signals and coordinate the global control strategy. Each computing unit fuses the generated local control signal with the control signals from other computing units to generate a global control signal:

[0130]

[0131] Among them, u glob Represents the global control signal, u loc represents the local control signal, α iis the weight of the ith computing unit, t i is the sampling time of the ith computing unit;

[0132] The global control signal is sent to the actuator of the inspection robot, which performs corresponding movements according to the control signal to realize the coordinated movement of the inspection robot; the actual motion data of the inspection robot is collected through sensors, the actual motion data is compared with the expected motion data, the control error is calculated, the control strategy is adjusted according to the control error, and a new coordinated control signal is generated.

[0133] S5: Send the coordinated control signal to the actuator of the inspection robot to realize the coordinated movement of the inspection robot.

[0134] Furthermore, the execution of the coordinated control signal includes that the actuator receives the control signal from the computing unit, and the control unit in the actuator is responsible for analyzing the signal and converting it into a drive signal for the motor and a control signal for the actuator;

[0135] The control unit in the actuator adjusts the movement of the motor and actuator according to the analyzed control signal. The motor control unit converts the signal into the current or speed control signal of the motor to accurately control the movement of each joint; the actuator control unit converts the signal into the force or displacement control signal of the actuator to accurately control the operation of the end effector.

[0136] The sensors on the inspection robot monitor the robot's motion status and external environment information in real time. The sensor data is fed back to the control unit of the actuator through the data acquisition system. The control unit evaluates the effect of the control signal based on the feedback data and uses it to adjust future control strategies.

[0137] Example 2

[0138] Reference Figure 2 and Figure 3 , which is an embodiment of the present invention, provides a fault diagnosis method for an inspection robot based on multi-sensor fusion. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0139] In order to verify the effectiveness of the inspection robot fault diagnosis method based on multi-sensor fusion provided by the present invention, an experiment was conducted on a simulated automobile engine assembly line. The inspection robot used in the experiment is a collaborative robot with seven degrees of freedom, which can simulate the flexible movement of human arms. Before the experiment, the joints of the robot were lubricated and maintained to ensure that its movement was smooth and noiseless. At the same time, the robot's sensors were calibrated to ensure the accuracy of the data.

[0140] Next, a multi-sensor fusion algorithm was integrated into the robot's control system. The algorithm combines machine learning models with dynamic system models to predict dynamic changes during the assembly of engine parts based on real-time data and previous experience. The adaptive algorithm can adjust the robot's motion trajectory and force control in real time based on the prediction results to adapt to different working conditions.

[0141] In the experiment, three different assembly tasks were designed, involving the engine piston assembly, connecting rod assembly and crankshaft assembly. Each task includes multiple steps such as grasping, positioning, alignment and fixing. When the robot performs the task, a high-precision laser scanner is used to measure the placement of parts to evaluate the assembly error. At the same time, a power meter is used to measure the energy consumption of the robot during the assembly process. The experimental data is shown in Table 1.

[0142] In addition, a high-speed camera was used to record the robot's performance of the task, and video analysis was used to evaluate the robot's motion smoothness and stability. A force sensor was also used to measure the force exerted by the robot during the assembly process to evaluate the accuracy of its force control.

[0143] Table 1 Experimental data analysis

[0144]

[0145]

[0146] It can be seen from the table that the inspection robot using the multi-sensor fusion fault diagnosis method has reduced the time required for each experimental task, significantly reduced the average assembly error, and reduced the average energy consumption. In addition, the robot's movement smoothness and force control accuracy have also been improved. These data show that the present invention significantly improves the performance of the inspection robot in complex industrial tasks in the embodiment.

[0147] By introducing prediction models and adaptive algorithms, the robot can better cope with model uncertainty and external interference, improving control accuracy and efficiency. At the same time, the synergy between inspection robots is taken into account, making the robots more coordinated when performing tasks and improving overall performance. By optimizing the adaptive algorithm and the adjustment strategy of control parameters, the computational complexity is reduced and the real-time performance is improved.

[0148] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0149] Example 3

[0150] The third embodiment of the present invention is different from the first two embodiments in that:

[0151] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0152] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0153] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0154] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0155] Example 4

[0156] Reference Figure 4 , is an embodiment of the present invention, and provides a patrol robot collaborative control system based on multi-sensor fusion, characterized in that it includes a data acquisition module, a prediction model module, a control strategy module, an actuator module, a communication module, and an adaptive algorithm module;

[0157] The data acquisition module collects the status data and external environment data of the inspection robot in real time.

[0158] The prediction model module establishes a prediction model, which predicts the future state of the inspection robot based on the current state and external environment information.

[0159] The control strategy module defines the control strategy, adjusts the control parameters according to the prediction error and the covariance matrix of the prediction error, and generates the collaborative control signal of the inspection robot.

[0160] The actuator module receives the collaborative control signal, adjusts the joint angle of the inspection robot, the posture and torque of the end effector, and realizes the collaborative movement of the inspection robot.

[0161] The communication module realizes network communication between computing units, ensures data synchronization and transmission of control signals, and coordinates the global control strategy.

[0162] The adaptive algorithm module adjusts the control parameters to achieve adaptive control according to the prediction error and the covariance matrix of the prediction error.

[0163] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A fault diagnosis method for inspection robots based on multi-sensor fusion, characterized in that: include, Collect the inspection robot's motion data and external environment information; Establish a prediction model for the inspection robot and generate prediction control signals based on the prediction model; According to the error between the predicted control signal and the actual control signal, the control parameters of the inspection robot are adjusted through an adaptive algorithm; Generate a collaborative control signal for the inspection robot according to the adjusted control parameters; The coordinated control signal is sent to the actuator of the inspection robot to realize the coordinated movement of the inspection robot.

2. The inspection robot fault diagnosis method based on multi-sensor fusion as claimed in claim 1, characterized in that: The collection of motion data and external environment information includes using sensors and monitoring equipment to collect motion data of the inspection robot, the motion data including position, velocity and acceleration, using force sensors to measure the torque and force at the joints, and the force of the end effector; at the same time, collecting external environment information, capturing the three-dimensional structure of the surrounding environment, obtaining pressure and shape information of contact with objects, and monitoring the ambient temperature and humidity; Appropriate sensors are installed on the joints and end effectors of the inspection robot, and the sensors are connected to the control unit; the sensor data is transmitted to the control unit in real time via wireless means, and the collected data is processed in the control unit.

3. The inspection robot fault diagnosis method based on multi-sensor fusion as claimed in claim 2, characterized in that: The establishment of the prediction model of the inspection robot includes selecting initial parameters and collecting data from the movement of the inspection robot; The data includes motion trajectory, force, torque, and external environment information. The objective function J(θ) is constructed based on the collected data: g k (θ)≤0,k=1,2,…,K Among them, y i represents the actual observed output value, represents the predicted output value, λ regularization parameter, μ j (θ,t) is the dynamic characteristic function, Δt represents the prediction time range, M represents the number of dynamic characteristic functions, N represents the number of data points, and g k (θ) represents the inequality constraint, k represents the number of inequality constraints; Ensure that the robot does not violate the laws of physics during movement, the force does not exceed the maximum bearing capacity of the actuator, and the joint angle does not exceed the range of motion. Set dynamic constraints: g1(θ)=min(F max -F(θ),0)≤0 g2(θ)=min(τ max -τ(θ),0)≤0 g3(θ)=min(γ max -γ(θ),0)≤0 Where F(θ) is the force calculated from the model parameters θ, and F max is the maximum allowable force, τ(θ) is the moment calculated from the model parameter θ, τ max is the maximum allowable moment, γ(θ) is the joint angle calculated from the model parameter θ, and γ max is the maximum allowed joint angle; To ensure that the robot's motion trajectory does not collide with itself or enter areas outside the workspace, set kinematic constraints: g4(θ)=min(d self (θ),0)≤0 g5(θ)=min(d ws (θ),0)≤0 Among them, d self (θ) represents self-collision detection, d ws (θ) represents the workspace boundary detection; Ensure that the robot's performance indicators when performing tasks meet specific requirements and set performance constraints: g6(θ)=T(θ)-T max ≤0 g7(θ)=E(θ)-E max ≤0 Among them, T(θ) represents the time required to complete the task, T max represents the maximum allowed time, E(θ) represents the energy consumption required to complete the task, and E max Indicates the maximum allowable energy consumption; To ensure smooth robot motion and avoid sudden acceleration or deceleration, set smoothness performance constraints: Among them, σ 2 Represents the upper limit of the smoothness indicator; To ensure that the two arms of the inspection robot maintain coordination during movement and avoid conflicts or uncoordinated movements, set coordination constraints: Among them, y left and right Represent the predicted outputs of the left and right arms, δ 2 Indicates the upper limit of the synergy index.

4. The inspection robot fault diagnosis method based on multi-sensor fusion as claimed in claim 3, characterized in that: Generating the prediction control signal includes predicting the state of the inspection robot in the future through a prediction model; According to the predicted future state, the control signal required to achieve the desired target state is calculated. The control signal includes the joint torque, joint angle target and the pose target of the end effector: u pred (t)=C(x pred (t),x des (t)) Among them, u pred (t) represents the predicted control signal generated at time t, C is the control strategy function, which calculates the control signal based on the predicted state and the desired state, x pred (t) represents the predicted robot state at time t, x des (t) represents the desired robot state at time t; The predicted control signal is sent to the robot's control system. At the same time, the control system generates an actual control signal based on the current sensor data. The actual control signal drives the motor and actuator to make the robot's actual state reach the predicted state: u act (t)=K(t)·(x des (t)-x act (t))+u pred (t) K(t)=P(t)C T (x des (t))C(x(t)) Among them, u act (t) represents the control signal actually applied at time t, x act (t) represents the state of the actual application at time t, K(t) represents the feedback gain matrix, P(t) represents the prediction error covariance matrix, C(x des (t)) represents the transition matrix of the desired state, and C(x(t)) represents the transition matrix of the actual state.

5. The inspection robot fault diagnosis method based on multi-sensor fusion as claimed in claim 4, characterized in that: The adaptive algorithm for adjusting control parameters includes calculating prediction errors; e(t)=x(t)-x pred (t) Among them, x(t) represents the actual state; A dynamically changing prediction error threshold Δe is set to determine whether the prediction error is within an acceptable range. When the prediction error e(t) exceeds the dynamic threshold Δe, the prediction error processing mechanism is started, and a dynamically changing prediction error covariance matrix threshold is set to determine whether the prediction error covariance matrix is ​​stable. Compute the forecast error covariance matrix: P(t)=Cov(x(t),x pred (t)) A dynamically changing prediction error covariance matrix threshold ΔP is set to determine whether the prediction error covariance matrix is ​​stable. When the change rate of the prediction error covariance matrix P(t) exceeds the dynamic threshold ΔP, the prediction error covariance matrix processing mechanism is started. According to the results of the prediction error processing module and the prediction error covariance matrix processing module, the control parameters are adjusted. The amplitude and frequency of adjusting the control parameters are proportional to the processing results of the prediction error and the prediction error covariance matrix.

6. A method for fault diagnosis of an inspection robot based on multi-sensor fusion as claimed in claim 5, characterized in that: Generating the coordinated control signal of the inspection robot includes distributing the control task to multiple computing units using distributed computing technology and coordinating the global control strategy through network communication; Decompose the control task into multiple subtasks. Each subtask is handled by a computing unit. The subtasks are reasonably allocated according to the computing power of the computing unit and the network delay. Each computing unit generates a local control signal based on the subtask and local sensor data. Through network communication, all computing units are ensured to have the same time reference and status information. The computing units communicate through the network to exchange control signals and coordinate the global control strategy. Each computing unit fuses the generated local control signal with the control signals from other computing units to generate a global control signal: Among them, u glob Represents the global control signal, u loc represents the local control signal, α i is the weight of the ith computing unit, t i is the sampling time of the ith computing unit; The global control signal is sent to the actuator of the inspection robot, which performs corresponding movements according to the control signal to realize the coordinated movement of the inspection robot; the actual motion data of the inspection robot is collected through sensors, the actual motion data is compared with the expected motion data, the control error is calculated, the control strategy is adjusted according to the control error, and a new coordinated control signal is generated.

7. The inspection robot fault diagnosis method based on multi-sensor fusion as claimed in claim 6, characterized in that: The execution of the coordinated control signal includes: the actuator receives the control signal from the computing unit, and the control unit in the actuator is responsible for analyzing the signal and converting it into a drive signal for the motor and a control signal for the actuator; The control unit in the actuator adjusts the movement of the motor and actuator according to the analyzed control signal. The motor control unit converts the signal into the current or speed control signal of the motor to accurately control the movement of each joint; the actuator control unit converts the signal into the force or displacement control signal of the actuator to accurately control the operation of the end effector. The sensors on the inspection robot monitor the robot's motion status and external environment information in real time. The sensor data is fed back to the control unit of the actuator through the data acquisition system. The control unit evaluates the effect of the control signal based on the feedback data and uses it to adjust future control strategies.

8. A system using a multi-sensor fusion-based inspection robot fault diagnosis method as claimed in any one of claims 1 to 7, characterized in that: It includes data acquisition module, prediction model module, control strategy module, actuator module, communication module and adaptive algorithm module; The data acquisition module collects the status data and external environment data of the inspection robot in real time; The prediction model module establishes a prediction model, which predicts the future state of the inspection robot based on the current state and external environment information; The control strategy module defines the control strategy, adjusts the control parameters according to the prediction error and the covariance matrix of the prediction error, and generates the collaborative control signal of the inspection robot; The actuator module receives the coordinated control signal, adjusts the joint angle of the inspection robot, the position and torque of the end effector, and realizes the coordinated movement of the inspection robot; The communication module implements network communication between computing units, ensures data synchronization and transmission of control signals, and coordinates global control strategies; The adaptive algorithm module adjusts the control parameters to realize adaptive control according to the prediction error and the covariance matrix of the prediction error.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images