A multi-robot trajectory fault detection method based on trajectory diffusion

Abstract

The application relates to a kind of multi-robot trajectory fault detection methods based on trajectory diffusion, comprising: each robot in the system is respectively equipped with sensor, and the position of itself is measured in real time through sensor and the position information of other robots is solved;Each robot uses respective measurement data to construct multi-robot state model in real time, the model is distributed partially observable Markov decision process, and the overall operation state of multi-robot system is described through multi-robot state model;Collect the observation state data set when the multi-robot system normally operates, build and train trajectory diffusion model;In the process of system operation, the real-time robot position data and historical data obtained are input into trajectory diffusion model, the possible trajectory of multi-robot system in the future is predicted, and real-time detection of fault robot is completed through error evaluation.Compared with prior art, the application reduces the dependence of the system on global information, and improves the adaptability and robustness of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-robot system fault detection, and in particular to a multi-robot trajectory fault detection method based on trajectory diffusion. Background Art

[0002] A multi-robot system is composed of multiple intelligent, autonomous, and collaborative robots. Through rational organization and collaboration, these robots can accomplish complex tasks that would be difficult for a single robot to accomplish. With the rapid development of robotics technology, multi-robot systems have found widespread application in a variety of fields, including industrial production, military equipment, and consumer services. However, as multi-robot systems grow in size and complexity, system failures are becoming increasingly prominent, becoming a key issue hindering their application.

[0003] Existing fault detection technologies for multi-robot systems mainly include the following three categories:

[0004] Model-based methods: Model-based fault detection methods build a mathematical model of the robot, describing its dynamics, control logic, and environmental characteristics. Furthermore, they detect faults in the system by comparing the residuals between the model's predictions and the actual observed values. If the residuals exceed a predefined threshold, this indicates a possible fault. Model-based methods can also detect faults by performing online estimation of system parameters. Discrepancies between the estimated and actual values ​​indicate a possible fault.

[0005] Data-based methods: Data-based fault detection methods use historical data to analyze the statistical characteristics of robot behavior and identify anomalies by examining the distribution and changing trends of the data. For robots with time-series characteristics, abnormal behavior can be identified by analyzing the temporal characteristics of the robot's sensor data, such as timing patterns and periodic changes. Signal processing techniques can also be used to filter and reduce noise from sensor data to extract effective information and improve fault detection accuracy.

[0006] Learning-based methods: Supervised learning-based fault detection methods utilize labeled data to train models, enabling them to identify different types of fault patterns. Unlike supervised learning, unsupervised learning methods do not require labeled data. Instead, they identify potential faults through techniques such as automatic data clustering and anomaly detection. If sample data is insufficient, reinforcement learning-based fault detection methods can be considered. This method learns optimal strategies through interaction with the environment and improves fault detection performance through trial and error.

[0007] It is worth noting that the operating environment of multi-robot systems is often complex and harsh, making it difficult to ensure high-quality wireless communication. This means that in most cases, robots can only rely on onboard sensor measurements to obtain local information. In addition, the decentralized nature of the system means that there is no central node to coordinate and plan the overall behavior of the system. Individual robots need to monitor abnormal conditions both themselves and their neighbors to ensure their own safe operation. Existing fault detection algorithms typically rely on a set of independent detectors to complete the fault detection task for themselves and their neighbors separately. As the number of robots in the system and the connections between them increase, the overall complexity of the algorithm will increase significantly. These shortcomings seriously restrict the application of existing multi-robot system fault detection algorithms in practical systems. Summary of the Invention

[0008] The purpose of the present invention is to provide a multi-robot trajectory fault detection method based on trajectory diffusion in order to overcome the defects of the above-mentioned prior art, that is, the existing multi-robot system learning-based fault detection algorithm can only identify the dynamic model of the system but cannot fit the overall fault trajectory; and the defects of high overall complexity and low performance of the fault detection algorithm when the number of robots and connection relationships in the system increase.

[0009] The purpose of the present invention can be achieved by the following technical solutions:

[0010] A multi-robot trajectory fault detection method based on trajectory diffusion includes the following steps:

[0011] Each robot in the multi-robot system is equipped with a sensor, which measures its own position in real time and calculates the position information of other robots;

[0012] Each robot uses its own measurement data to build a multi-robot state model in real time. The multi-robot state model is a distributed partially observable Markov decision process that describes the overall operating state of the multi-robot system.

[0013] Collecting an observation state dataset of the multi-robot system during normal operation, and building and training a trajectory diffusion model;

[0014] During the operation of the multi-robot system, the acquired real-time robot pose data and historical data are input into the trained trajectory diffusion model to predict the possible future trajectories of the multi-robot system, and the real-time detection of faulty robots is completed through error evaluation.

[0015] Furthermore, the multi-robot state model determines the motion state of the multi-robot system at the current moment by the control input and the motion state of the robots at the previous moment;

[0016] The control input is determined by a system control strategy based on a distributed partially observable Markov decision process.

[0017] Furthermore, the multi-robot state model is described as a nine-tuple G=<N,S,A,T,Ω,O,r,U,γ> ;

[0018] Where N is the total number of robots in the multi-robot system, S is the state space of the robot system, and A=× i A i is the set of joint actions, A i is the set of behaviors of robot i, × i A i For each A i Multiply them in sequence, T is the set of conditional transition probabilities between states, Ω=× i Ω i is the set of joint observations, Ω i is the observation set of robot i, × i Ω i For each Ω i Multiply them in sequence, O is the set of conditional transition probabilities between states, r is the shared reward function r(s,a), s,a are the system state and behavior at the current moment, U is the initial state distribution at t=0, and γ∈[0,1] is the discount factor of the system decision.

[0019] Furthermore, the multi-robot state model describes the state of each robot as a state parameter in discrete time, and samples the initial global state s0∈S from the distribution U;

[0020] At each time sampling point t, each robot i only observes one local measurement information The measurement information is generated by the function O(s,a)∶S×A→Ω, and a behavior a∈A is selected according to the corresponding control strategy. The joint action of all robots is a∈A=A1×A2×...×A N The multi-robot state model is transferred to the next state s′ according to the state transfer function P(s′|s,a):S×A→S to obtain the overall operating state of the multi-robot system.

[0021] Furthermore, the observation state data set D of the multi-robot system during normal operation is composed of multiple groups of robot operation trajectories, each robot operation trajectory The observation value of the i-th robot at time t Build;

[0022] The observation state dataset D during normal operation of the multi-robot system is a static dataset obtained in an offline manner.

[0023] Furthermore, the trajectory diffusion model is a generative model based on the diffusion process, which gradually diffuses the data from the clean signal x0 to the Gaussian noise x T , and then from the Gaussian noise x T Gradually learn to restore the data to the original signal x0;

[0024] The input of the trajectory diffusion model is the trajectory of each robot The trajectory diffusion model is used to learn the optimal trajectory data sampled from the noise as the prediction result of the trajectory at the future moment.

[0025] Furthermore, when predicting the trajectory at a future moment, the trajectory diffusion model also uses the historical trajectory information of the current robot as a condition through a conditional guidance module to guide the sampling degree of the trajectory diffusion model, so that the trajectory generated by the trajectory diffusion model is integrated with the actual trajectory information before the current moment.

[0026] Furthermore, the process of predicting the possible future trajectories of the multi-robot system specifically includes:

[0027] Taking the position information of each robot as the starting point, the position information of other robots collected by the robot's sensor is converted into corresponding position information observation values ​​through posture solution; the position information observation values ​​are used as the conditional input of the initial Gaussian noise data, and the trajectory diffusion model is used for iterative diffusion calculation to realize the sampling of the state trajectory of the entire multi-robot system based on the local observation of the robot, and the trajectory of the current multi-robot system at the future p-th time is obtained.

[0028] Furthermore, the calculation expression corresponding to the iterative diffusion calculation is:

[0029]

[0030] Where, is the initial trajectory noise of the i-th robot in the diffuse noise T time step, is the actual historical observation trajectory information of the i-th robot at time t-1, N is the total number of robots in the multi-robot system, is the predicted trajectory of the i-th robot at the c-th moment in the future.

[0031] Furthermore, the error assessment method is specifically as follows:

[0032] Compare the predicted possible future trajectory of the multi-robot system with the actual observed trajectory of the multi-robot system at the corresponding moment, calculate the residual value of each robot trajectory, and compare it with the threshold function θ th Compare and determine whether each robot has failed;

[0033] The calculation expression of the residual value is:

[0034]

[0035] Where, is the residual value of robot No. i at time t+p, is the predicted trajectory value of robot No. i at time t+p, is the actual observed value of the trajectory of robot No. i at time t+p;

[0036] The specific process of determining whether each robot has a fault is as follows:

[0037] when When , it is judged that the robot No. i deviates from the reasonable trajectory route and fails;

[0038] Otherwise, it is judged that the robot No. i is running on a reasonable trajectory and no fault occurs.

[0039] Compared with the prior art, the present invention has the following advantages:

[0040] (1) Traditional multi-robot fault detection algorithms can usually only perform fault diagnosis based on the system's dynamic model and cannot effectively fit the overall fault trajectory. By applying the trajectory diffusion method to multi-robot fault detection, this invention can perform trajectory prediction and fault detection using actual observation data and historical data without an accurate dynamic model, thus overcoming the shortcomings of traditional methods in fault trajectory fitting.

[0041] (2) This invention employs a fault detection method based on local pose information. A robot uses position sensors (such as lidar or cameras) to obtain real-time pose information of itself and its neighbors, without relying on complex communication networks or information transmission mechanisms. Because this method does not rely on wireless communication networks, it is suitable for operation in environments with limited or no stable communication networks, improving the adaptability and robustness of the system.

[0042] (3) This paper uses a distributed partially observable Markov decision process (Dec-POMDP) ​​to construct a state model for a multi-robot system, allowing each robot to make decisions based solely on its own local observations without requiring state sharing or external input across the team. This distributed modeling approach reduces the system's reliance on global information, enabling the fault detection system to more efficiently cope with the dynamic changes and uncertainties in the multi-robot system.

[0043] (4) The present invention applies a trajectory diffusion model, which generates trajectories based on a diffusion process, can recover data from noise, and generate optimal trajectories for future moments. This generative model, combined with historical trajectory information as conditional input, can efficiently generate future trajectory predictions and accurately determine whether the robot has malfunctioned. Compared with traditional methods, the trajectory diffusion model provides more accurate and flexible fault detection capabilities and can adapt to different operating environments in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 Schematic diagram of a flow chart of a multi-robot trajectory fault detection method based on trajectory diffusion provided in an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of the initial state of a particle world simulation environment provided in an embodiment of the present invention, including a multi-robot system consisting of three landmarks and three robots in initial positions;

[0046] Figure 3 The actual running trajectory of a multi-robot system in a simulation environment and the predicted trajectory generated by the trajectory diffusion model are provided in an embodiment of the present invention;

[0047] Figure 4 A schematic diagram of a residual detection result provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0049] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0050] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0051] Example 1

[0052] like Figure 1As shown, this embodiment provides a multi-robot trajectory fault detection method based on trajectory diffusion, comprising the following steps:

[0053] S1: Each robot in the multi-robot system is equipped with a sensor, which measures its own position in real time and calculates the position information of other robots;

[0054] S2: Each robot uses its own measurement data to build a multi-robot state model in real time. The multi-robot state model is a distributed partially observable Markov decision process that describes the overall operating state of the multi-robot system.

[0055] S3: Collect observation status datasets during normal operation of the multi-robot system, build and train trajectory diffusion models;

[0056] S4: During the operation of the multi-robot system, the acquired real-time robot pose data and historical data are input into the trained trajectory diffusion model to predict the possible future trajectories of the multi-robot system and complete the real-time detection of faulty robots through error evaluation.

[0057] Specifically, in step S1, the robot is equipped with sensors including a lidar, an odometer, and an inertial measurement unit. Upon system startup, these sensors activate synchronously. The odometer and inertial measurement unit obtain real-time position measurement data for the robot itself, while the lidar acquires distance information from other robots. This data is then transmitted to the robot controller for real-time position calculation, obtaining the local position information of other robots. Because this solution does not rely on wireless communication networks, it can be applied to most practical scenarios where communication is limited.

[0058] In step S2, the multi-robot state model constructs the current motion state of the robot system. The current motion state of the robot system is determined by both the control input and the robot's motion state at the previous moment. The robot's control input is determined solely by the system control strategy based on a distributed partially observable Markov decision process (Dec-POMDP). In Dec-POMDP, the system state is determined solely by the independent observation information of each robot. Therefore, the overall fault detection reference model does not rely on external input or team information sharing, requiring only observation information from local sensors.

[0059] The multi-robot state model is described as a distributed partially observable Markov decision process, which is represented by a nine-tuple G =<N,S,A,T,Ω,O,r,U,γ> ,in:

[0060] N is the total number of robots in the system;

[0061] S is the state space of the robot system;

[0062] A i is the set of behaviors of robot i, A=× i A i It is a collection of joint actions;

[0063] T is the set of conditional transition probabilities between states, T(s,a,s′)=P(s′|s,a), where s,a are the system state and behavior at the current moment, and s′ is the state at the next moment;

[0064] Ω i is the observation set of robot i, Ω=× i Ω i is the set of joint observations;

[0065] O is the set of conditional transition probabilities between states;

[0066] r is the shared reward function r(s,a);

[0067] U is the initial state distribution at t = 0, from which the initial state of the system can be sampled;

[0068] γ∈[0,1] is the discount factor for system decision making.

[0069] The overall multi-robot system state model includes N robots {1, 2, ..., N}, whose states are described as state parameters in discrete time, and the initial global state s0∈S is sampled from the distribution U. At each time sampling point t, each robot i only observes a local measurement information The measurement information is generated by the function O(s,a)∶S×A→Ω, and a behavior a∈A is selected according to the corresponding control strategy. The joint action of all robots is a∈A=A1×A2×...×A N The system will transfer to the next state s′ according to the state transition function P(s′|s,a):S×A→S.

[0070] In step S3, the normal operation dataset of the multi-robot system is a static dataset obtained offline, without collecting data from the online environment. The data in the dataset consists of multiple groups of multi-robot system operation trajectories, where the trajectories The observation value of the i-th robot at time t Build.

[0071] The trajectory diffusion model is a generative model based on the diffusion process, which gradually diffuses the data from the clean signal x0 to the Gaussian noise x T , and then from the Gaussian noise x T Gradually learn to restore the data to the original signal x0. This model is applied to trajectory prediction, which is to convert the trajectories of multiple robots in a certain discrete time. As data, the model learns to sample optimal trajectory data from noise as predictions for future trajectories. To integrate the generated trajectories with the actual trajectory information before the current moment, a conditional guidance module is applied to the trajectory diffusion model. This module can input historical trajectory information into the model as a condition, guiding the model's sampling level and improving the quality of the generated model.

[0072] In step S4, each robot uses its own position information as the starting point and uses the distance information of other robots collected by the measurement sensor to calculate the corresponding position information observation value through the posture solution. These observation values ​​are used as the conditional input of the initial Gaussian noise data. Through the iterative diffusion step, the state trajectory of the entire multi-robot system is sampled based on the local observation of the robot. The sampling process is as follows, where represents the initial trajectory noise of the i-th robot in the diffuse noise T time step, Represents the actual historical observation trajectory information of the i-th robot at time t-1:

[0073]

[0074] Where, is the initial trajectory noise of the i-th robot in the diffuse noise T time step, is the actual historical observation trajectory information of the i-th robot at time t-1, N is the total number of robots in the multi-robot system, is the predicted trajectory of the i-th robot at the c-th moment in the future.

[0075] The predicted trajectory at the pth moment in the future generated by the trajectory diffusion model It shows the optimal trajectory of the current multi-robot system at the future time p, which can be used as the reference trajectory at the time t+p, compared with the actual observation value at the time t+p and the residual value calculated:

[0076]

[0077] For the calculated residual value, the threshold function θ is designed th ,when When , it is judged that the robot No. i deviates from the reasonable trajectory route and fails; when When , it is judged that the robot No. i is running on a reasonable trajectory and no fault occurs.

[0078] Specific implementation process:

[0079] The following uses the multiagent-praticle-envs simulation environment as an example. The multiagent-praticle-envs simulation environment is a multi-agent simulation environment that can simulate various interactions between agents. This embodiment uses the mpe-spread task to illustrate the present invention. The task goal is to ensure that all robots in a three-robot system can reach a landmark location without conflict. The specific fault detection method includes the following steps:

[0080] Step 1: The robot node in the simulation environment can simulate the robot measurement sensor, obtain the robot's own position and speed information through the simulation observation space, and obtain the relative position information of other robots in real time. Figure 2 As shown, the host robot can obtain its own position information in real time, and at the same time obtain the relative position information of other robots and all landmarks. After solving, the absolute position information of all robots and landmarks can be obtained.

[0081] Step 2: Start the simulation environment. When the algorithm of the present invention is offline, collect a static dataset of the fault-free robot system under the current mpe-spread task, extract the multi-robot system operation trajectory in the dataset, build a multi-robot state model and a joint observation set Ω, and train the trajectory diffusion model.

[0082] Step 3: Restart the simulation environment and start all robot nodes, numbered as [0, 1, 2]. At this time, the algorithm of the present invention is running online, and the multi-robot system runs according to its control strategy. The running trajectory is Figure 2 At the same time, robot 0, as the host robot, obtains the observed position information of itself and other robots. Referring to formula (1), this information is input into the trajectory diffusion model as the historical observation trajectory information. The trajectory diffusion model is used to infer the predicted future trajectory, as shown in the following example: Figure 3 Indicated by the dotted line.

[0083] Step 4: During the operation of the multi-robot system, refer to formula (2) and calculate the residual value in real time based on the observed position information of each robot observed by the host robot and the future trajectory information predicted at the previous moment.

[0084] Step 5: Design threshold function θ th In this embodiment, take θ th =0.06, it is considered that when the residual value between the actual trajectory of the robot and the predicted trajectory is greater than the threshold, the robot is considered to have failed and the trajectory has deviated.

[0085] Step 6: Figure 3As shown in the figure, the residual values ​​of the actual trajectory and the predicted trajectory of each robot calculated by the host robot No. 0 are visualized during the operation, and the threshold function is set. Figure 3 It can be seen intuitively that the robot No. 2 has a fault. In the first 5 moments, it deviates significantly from the optimal route to the target, and in the 15th to 20th moments, it moves significantly near the target landmark. This fault is reflected in Figure 4 During the residual detection, if the trajectory residual value of the robot exceeds the set threshold, the host robot will issue a fault alarm and try to adjust the control strategy of the multi-robot system.

[0086] This invention provides a multi-robot trajectory fault detection method based on trajectory diffusion. Compared with existing methods, it has the following significant innovations: First, traditional multi-robot fault detection algorithms typically perform fault diagnosis based solely on the system's dynamic model and are unable to effectively fit the overall fault trajectory. By applying the trajectory diffusion method to multi-robot fault detection, this invention can perform trajectory prediction and fault detection using actual observation data and historical data, even without an accurate dynamic model. This overcomes the shortcomings of traditional methods in fitting fault trajectories. Second, the invention adopts a fault detection method based on local pose information. Robots obtain their own and their neighbors' pose information in real time through position sensors (such as lidar or cameras), without relying on complex communication networks or information transmission mechanisms. Because this method does not rely on wireless communication networks, it is suitable for operation in environments with limited or no stable communication networks, improving the system's adaptability and robustness. Third, the invention utilizes a distributed partially observable Markov decision process (Dec-POMDP) ​​to construct a state model for the multi-robot system, allowing each robot to make decisions based solely on its own local observation information, without requiring state sharing or external input across the team. This distributed modeling approach reduces the system's reliance on global information, enabling the fault detection system to more efficiently cope with the dynamic changes and uncertainties in multi-robot systems. Finally, the present invention utilizes a trajectory diffusion model, which generates trajectories based on a diffusion process, capable of recovering data from noise and generating optimal trajectories for future moments. This generative model, incorporating historical trajectory information as conditional input, efficiently generates future trajectory predictions and accurately determines whether a robot has failed. Compared to traditional methods, the trajectory diffusion model provides more accurate and flexible fault detection capabilities, capable of adapting to diverse operating environments in real time.

[0087] In summary, the present invention realizes trajectory fault detection for multiple robots by acquiring only local posture information, reducing the system's dependence on global information through distributed modeling, and applying trajectory diffusion models. It has broad application prospects and important social significance.

[0088] The preferred embodiments of the present application have been described above in detail. It should be understood that modifications and variations to the preferred embodiments could be made by those skilled in the art in light of the teachings above. It is therefore contemplated that the application can encompass other variations and modifications that fall within the scope of the claims.

Claims

1. A multi-robot trajectory fault detection method based on trajectory diffusion, characterized in that: The following steps are involved: Each robot in the multi-robot system is equipped with a sensor, which measures its own position in real time and calculates the position information of other robots; Each robot uses its own measurement data to build a multi-robot state model in real time. The multi-robot state model is a distributed partially observable Markov decision process that describes the overall operating state of the multi-robot system. Collecting an observation state dataset of the multi-robot system during normal operation, and building and training a trajectory diffusion model; During the operation of the multi-robot system, the acquired real-time robot pose data and historical data are input into the trained trajectory diffusion model to predict the possible future trajectory of the multi-robot system and to detect faulty robots in real time through error evaluation. The multi-robot state model determines the motion state of the multi-robot system at the current moment by the control input and the motion state of the robot at the previous moment; The control input is determined by a system control strategy based on a distributed partially observable Markov decision process; The multi-robot state model is described as a nine-tuple ; in, is the total number of robots in the multi-robot system, is the state space of the robot system, is a collection of joint actions, It's a robot The set of behaviors, For each Multiply them in sequence, is the set of conditional transition probabilities between states, is the set of joint observations, It's a robot The set of observations, For each Multiply them in sequence, is the set of conditional transition probabilities between states, is the shared reward function , is the system state and behavior at the current moment, is the initial state distribution at t=0, is the discount factor for system decision; The trajectory diffusion model is a generative model based on the diffusion process, which transforms the data from the clean signal Gradually diffuse into Gaussian noise , and then from Gaussian noise Gradually learn to restore the data to the original signal ; The input of the trajectory diffusion model is the trajectory of each robot The trajectory diffusion model is used to learn the optimal trajectory data sampled from the noise as the prediction result of the trajectory at the future moment; In the process of predicting the trajectory at future moments, the trajectory diffusion model also uses the historical trajectory information of the current robot as a condition through the conditional guidance module to guide the sampling degree of the trajectory diffusion model, so that the trajectory generated by the trajectory diffusion model is integrated with the actual trajectory information before the current moment.

2. A multi-robot trajectory fault detection method based on trajectory diffusion according to claim 1, characterized in that: The multi-robot state model describes the state of each robot as a state parameter in discrete time and distributes the state of each robot into a discrete state parameter. Sampling the initial global state ; At each time sampling point When each robot Only one local measurement information is observed , the measurement information is provided by the function Generate and select a behavior according to the corresponding control strategy , the joint action of all robots Make the multi-robot state model based on the state transition function Transition to the next state , and obtain the overall operating status of the multi-robot system.

3. The multi-robot trajectory fault detection method based on trajectory diffusion according to claim 1 is characterized in that: The observation state dataset of the multi-robot system during normal operation It is composed of multiple groups of robot running tracks, each robot running track By A robot in Local measurement information at the moment Build; The observation state dataset of the multi-robot system during normal operation It is a static dataset obtained offline.

4. The multi-robot trajectory fault detection method based on trajectory diffusion according to claim 1 is characterized in that: The process of predicting the possible future trajectories of the multi-robot system specifically includes: Taking the position information of each robot as the starting point, the position information of other robots collected by the robot's sensor is converted into the corresponding position information observation value through posture solution; the position information observation value is used as the conditional input of the initial Gaussian noise data, and the trajectory diffusion model is used for iterative diffusion calculation to realize the sampling of the state trajectory of the entire multi-robot system based on the local observation of the robot, and the state trajectory of the current multi-robot system in the future is obtained. The trajectory of time.

5. The multi-robot trajectory fault detection method based on trajectory diffusion according to claim 4 is characterized in that: The calculation expression corresponding to the iterative diffusion calculation is: Where, For the The initial trajectory noise of a robot in the diffuse noise T time step, For the The actual historical observation trajectory information of each robot at time t-1, is the total number of robots in the multi-robot system, For the future Moment The predicted trajectory of a robot.

6. The multi-robot trajectory fault detection method based on trajectory diffusion according to claim 1 is characterized in that: The error evaluation method is specifically as follows: Compare the predicted possible future trajectory of the multi-robot system with the actual observed trajectory of the multi-robot system at the corresponding moment, calculate the residual value of each robot trajectory, and compare it with the threshold function Compare and determine whether each robot has failed; The calculation expression of the residual value is: Where, For the Moment The residual value of the robot, For the Moment The predicted trajectory of each robot, For the Moment The actual observed value of the robot's trajectory; The specific process of determining whether each robot has a fault is as follows: when When judging A robot deviates from the reasonable trajectory and malfunctions; Otherwise, judge The robots ran on reasonable trajectories without any malfunctions.

Images