A Real-Time Control Method and Related Device for Cavity Robots Based on Dynamic Safety Boundaries

Abstract

This application discloses a real-time control method and related apparatus for cavity robots based on dynamic safety boundaries. The method includes: establishing a dynamic safety rule base; activating corresponding constraints in real time based on tissue characteristics obtained from sensor fusion; converting the fused state information into symbolic trajectories in discrete time steps, and using a time automaton to detect online whether the current safety constraints are met; if not, sampling to generate a control sequence under a model predictive control framework, predicting the trajectory using a physics engine, and then verifying it point-by-point using safety rules, retaining the successful trajectories and selecting the one with the highest score as the new motion path; parsing the path into a velocity command, driving the robot to execute it after impedance control loop and Lie group projection. This invention significantly reduces the risk of intraoperative collisions and improves the safety and real-time performance of cavity robot surgery through dynamic updating and closed-loop verification of safety rules.

Description

Technical Field

[0001] This invention relates to the fields of medical devices, artificial intelligence and robotics, specifically a real-time control method and related device for cavity robots based on dynamic safety boundaries. Background Technology

[0002] Surgical robots for intracavitary procedures have been widely used in minimally invasive interventional fields such as digestive, urological, and respiratory surgery. However, existing systems generally employ a rigid "offline planning-online tracking" architecture: the path is planned once before surgery based on static images, with only limited corrections made during the operation. When tissues deform in real time due to temperature, contact force, or instrument traction, the pre-set safety boundaries quickly fail, leading to complications such as instrument perforation of the cavity wall, thermal damage, or navigation drift. Simultaneously, heterogeneous multimodal sensor data and inconsistent coordinate systems make it difficult for the robot to complete state fusion and decision updates within milliseconds, further amplifying the risk of collisions. Traditional obstacle avoidance algorithms based on rules or simple potential fields lack a dynamic understanding of tissue mechanics, and path recalculation is time-consuming, making it difficult to meet the high-frequency safety constraints within narrow and tortuous cavities. Therefore, the industry urgently needs a closed-loop control method that can perceive tissue changes in real time, dynamically reconstruct safety boundaries, and instantaneously replan the motion path to improve surgical safety and operational smoothness. Summary of the Invention

[0003] To address the aforementioned technical problems, the present invention relates to a real-time control method and related apparatus for cavity robots based on dynamic safety boundaries, including but not limited to a real-time control device for cavity robots based on dynamic safety boundaries, electronic devices, computer-readable storage media, and computer program products.

[0004] Firstly, a real-time control method for cavity robots based on dynamic safety boundaries is provided, comprising the following steps:

[0005] S1: Establish a dynamic security rule base, which stores security constraint rules defined based on linear time-series logic; the security constraint rules can be adjusted by changes in organizational characteristics sensed by operating instructions or sensors;

[0006] S2: Acquire sensor data from the cavity robot, perform fusion processing on the sensor data to obtain fusion status information; the dynamic safety rule base activates the safety constraint rules of the corresponding subset as the current safety constraint rules based on the fusion status information;

[0007] S3: Based on the fusion state information and the current safety constraint rules, determine whether the motion of the cavity robot satisfies the current safety constraint rules;

[0008] S4: When the judgment result is not satisfied, based on the fusion state information, task target information and the current security constraint rules, generate a motion path that satisfies the current security constraint rules, and select one of them as a new motion path;

[0009] S5: Control the cavity robot to execute the new motion path.

[0010] In any embodiment of this application, step S1 includes:

[0011] A set of first-order linear temporal logic formulas is constructed, wherein the set of formulas takes the elastic modulus of cavity tissue, wall thickness, temperature and contact force as atomic propositions;

[0012] The set of formulas is compiled into a finite state monitor using a runtime verification framework. The finite state monitor takes an event stream as input and outputs a Boolean flag.

[0013] The finite state monitor is registered to the rule update interface, which triggers the finite state monitor to be recompiled when it receives an organizational characteristic change event.

[0014] The recompilation includes invoking a syntax rewriting engine, which inserts or deletes atomic propositions while maintaining the semantic equivalence of the set of formulas.

[0015] In any embodiment of this application, step S2 includes:

[0016] Simultaneous acquisition of intracavitary images, fiber Bragg grating strain, electromagnetic tracking pose, and micro-ultrasonic echoes via a multimodal sensor array;

[0017] The intracavitary image is input into a feature encoding network based on self-supervised learning to obtain a dense geometric feature tensor.

[0018] The fiber Bragg grating strain, the electromagnetic tracking pose, and the micro-ultrasonic echo are input into a cross-domain alignment module based on a graph neural network, and the output is a state node vector in a unified coordinate system.

[0019] Extended Kalman consistency filtering is performed on the state node vector to obtain the fused state information;

[0020] The dynamic security rule base listens to the characteristic change fields in the fusion status information and indexes the matching subset rules based on the field values.

[0021] In any embodiment of this application, step S3 includes:

[0022] Using discrete time steps as an index, the fused state information is converted into symbolic trajectories;

[0023] Transform the current security constraint rules into a time automaton;

[0024] The time automaton is executed to perform online model detection of the symbol trajectory, and a binary label indicating whether the condition is met or not is output.

[0025] When the condition is marked as unsatisfied, the time step of the first violation and the corresponding state variable are recorded.

[0026] In any embodiment of this application, step S4 includes:

[0027] Construct a motion tree with the fused state information as the root node, wherein the motion tree nodes store state, depth, and control input;

[0028] During the motion tree expansion phase, a sampling strategy based on model predictive control is invoked, which iteratively generates candidate control sequences within the prediction window;

[0029] The candidate control sequence is input into the physics engine, and the physics engine outputs the predicted trajectory;

[0030] The predicted trajectory is verified point by point using the current security constraint rules, and the verified trajectories are retained;

[0031] Run a path planning algorithm on the retained set of trajectories to obtain candidate paths and path scores;

[0032] The candidate path with the highest score is selected as the new motion path.

[0033] In any embodiment of this application, step S5 includes:

[0034] The new motion path is parsed into a speed command sequence;

[0035] The speed command sequence is input into the impedance control loop, and the impedance control loop modifies the stiffness matrix according to the real-time torque feedback to obtain the correction stiffness matrix.

[0036] By projecting the corrected stiffness matrix onto a Lie group, an executable control quantity on the constrained manifold can be obtained.

[0037] The executable control quantity is transmitted to the servo driver to drive the cavity robot to perform motion.

[0038] Secondly, a real-time control device for cavity robots based on dynamic safety boundaries is provided, including:

[0039] Security boundary storage unit: used to establish a dynamic security rule base, which stores security constraint rules defined based on linear time-series logic; the security constraint rules can be adjusted by changes in organizational characteristics sensed by operating instructions or sensors;

[0040] The perception and rule triggering unit is used to acquire sensor data from the cavity robot, perform fusion processing on the sensor data to obtain fusion state information, and activate the corresponding subset of safety constraint rules as the current safety constraint rules based on the fusion state information.

[0041] Rule discrimination unit: used to determine whether the motion of the cavity robot satisfies the current safety constraint rule based on the fusion state information and the current safety constraint rule;

[0042] Motion replanning unit: When the judgment result is not satisfied, it generates a motion path that satisfies the current safety constraint rules based on the fused state information, task target information and the current safety constraint rules, and selects one of them as a new motion path;

[0043] Replanning execution unit: used to control the cavity robot to execute the new motion path.

[0044] Thirdly, an electronic device is provided, comprising: a processor, a communication module, a sensor, a user interface, and a storage unit, wherein the storage unit is used to store computer program code, the program code including computer instructions. When the processor executes these instructions, the electronic device performs the methods described in the second aspect above and any of its embodiments.

[0045] Fourthly, another electronic device is provided, comprising: a processor, a wireless communication module, a touchscreen, a speaker, and a storage unit, wherein the storage unit is used to store computer program code, the program code including computer instructions. When the processor executes these instructions, the electronic device performs the methods described in the second aspect above and any of its embodiments.

[0046] Fifthly, a computer-readable storage medium is provided, wherein a computer program is stored, the program comprising program instructions. When these instructions are executed by a processor, the processor performs the methods described in the second aspect above and any of its embodiments.

[0047] In a sixth aspect, a computer program product is provided, the computer program product comprising a computer program or instructions. When the computer program or instructions are executed on a computer, the computer will perform the methods described in the second aspect above and any of its embodiments.

[0048] In this application, compared with the prior art, the present invention provides a method and related device for real-time monitoring of cavity environment based on multimodal perception. By establishing a dynamic safety rule base, the safety constraints described by linear temporal logic are transformed into a recompilable finite state monitor, enabling the rules to be adjusted online during surgery according to changes in tissue elasticity, wall thickness, temperature, and contact force. Subsequently, multimodal data from fiber Bragg gratings, electromagnetic tracking, micro-ultrasound, and intracavitary images are used, and after self-supervised feature encoding and cross-domain alignment with graph neural networks, fused state information is obtained through extended Kalman consistency filtering. This information drives the rule base to activate corresponding subsets in real time, and online model detection of symbol trajectories is performed using a time automaton. If a violation is found, the model prediction control sampling strategy is immediately invoked within the motion tree framework, and the trajectory is predicted by the physics engine and verified point by point with the current safety constraints. The compliant path is retained, and the optimal one is selected as the new motion path. The path is used to generate servo commands through impedance control and Lie group projection, realizing closed-loop safe movement of the cavity robot in the dynamic environment. Compared with traditional static boundary and offline planning schemes, the present invention fully couples perception, rule updating, path recalculation, and execution control, significantly improving surgical safety and operational smoothness. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the background art, the accompanying drawings used in the embodiments of this application or the background art will be described below.

[0050] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the specification, serve to illustrate the technical solutions of this application.

[0051] Figure 1 This is a schematic diagram of a real-time control method for a cavity robot based on dynamic safety boundaries, as proposed in an embodiment of this application.

[0052] Figure 2 This is a schematic diagram of a real-time control device for a cavity robot based on dynamic safety boundaries, as proposed in an embodiment of this application.

[0053] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0054] To enable those skilled in the art to more fully understand the technical solutions of this application, the technical solutions of this application will be explained in detail and clearly with reference to the accompanying drawings. It should be particularly noted that the described embodiments are only some examples of this application and do not represent all of them. Based on these embodiments, those skilled in the art can directly deduce all other possible implementation schemes without creative thinking, and these are also included within the protection scope of this application.

[0055] In the specification, claims, and related drawings of this application, the terms "first," "second," etc., are used only to distinguish different elements and do not imply any specific order. Furthermore, the use of "comprising" and "having," and their variations, indicates non-exclusive inclusion. This means that if a process, method, system, product, or device comprises a series of steps or components, it indicates that the process, method, system, product, or device is not limited to the listed steps or components and may also include other steps or components not listed, or other inherent steps or units thereof.

[0056] The term "embodiment" as used herein refers to any instance combining a particular feature, structure, or characteristic, which may be at least one embodiment of this application. The "embodiments" mentioned herein do not necessarily refer to the same specific case, nor do they imply that they are independent or exclusive alternatives. Those skilled in the art will understand that the embodiments described herein can be used with other embodiments. It should be clarified that in this application, "at least one" includes one or more instances, "multiple" means two or more instances, and "at least two" refers to two or more instances.

[0057] It should be understood that the method embodiments of this application can also be implemented by a processor executing computer program code. The embodiments of this application will now be described with reference to the accompanying drawings.

[0058] Please see Figure 1 , Figure 1 This is a schematic diagram of a real-time control method for a cavity robot based on dynamic safety boundaries, as proposed in an embodiment of this application.

[0059] 101. Construct a dynamic security rule base: Establish a dynamic security rule base, which stores security constraint rules defined based on linear time-series logic; the security constraint rules can be adjusted by changes in organizational characteristics sensed by operating instructions or sensors.

[0060] In this embodiment, the safety constraint rules include a set of first-order linear sequential logic formulas, with the elastic modulus of the cavity tissue, wall thickness, temperature, and contact force as atomic propositions.

[0061] In this embodiment, a runtime verification framework is used to compile the set of formulas into a finite state monitor, which takes an event stream as input and outputs a Boolean flag.

[0062] In this embodiment, the change adjustment includes: registering the finite state monitor to the rule update interface, wherein the rule update interface triggers the finite state monitor to recompile when it receives an organizational characteristic change event; the recompiling includes calling the syntax rewriting engine, wherein the engine inserts or deletes atomic propositions while maintaining the semantic equivalence of the formula set.

[0063] In this embodiment, the set of formulas can also include other information as atomic propositions, such as the near-infrared absorption coefficient, impedance spectrum, fluorescence intensity, and shear wave elastography parameters of cavity tissues.

[0064] 102. Data Fusion and Rule Activation: Acquire sensor data from the cavity robot, perform fusion processing on the sensor data to obtain fusion status information; the dynamic safety rule base activates the safety constraint rules of the corresponding subset as the current safety constraint rules based on the fusion status information.

[0065] In this embodiment, the acquisition of sensor data from the cavity robot and the fusion processing of the sensor data to obtain fused state information are based on the patent "A method and related device for real-time monitoring of cavity environment based on multimodal perception" (Publication No.: CN120597218A), the patent "A method and related device for cavity multimodal detection based on transfer learning and incremental learning" (Publication No.: CN120654203A), or the patent "An adaptive fusion method and related device for multi-source sensing data for flexible surgical instruments" (Publication No.: CN120597219A).

[0066] In another possible implementation, the acquisition of sensor data from the cavity robot and the fusion processing of the sensor data to obtain fused state information can be achieved using other methods.

[0067] 103. Real-time verification of motion safety: Based on the fusion state information and the current safety constraint rules, determine whether the motion of the cavity robot meets the current safety constraint rules.

[0068] In this embodiment, using discrete time steps as indices, the fused state information is transformed into symbolic trajectories, and the current security constraint rules are transformed into time automata; the time automata is then used to perform online model detection on the symbolic trajectories, outputting binary labels indicating whether the conditions are met or not; when the label indicates that the conditions are not met, the time step in which the violation first occurs and the corresponding state variable are recorded.

[0069] In another possible implementation, the symbolic trajectory can be replaced with a continuous differential containing interval trajectory, and the security constraint rules are expressed as signal timing logic formulas; interval reachability analysis and the Zonotope envelope propagation algorithm are used to quickly verify the trajectory through over-approximation; if a violation is found, a refinement strategy guided by counterexamples is used to locally tighten the interval within the critical time window, and the input set boundary that caused the violation is recorded.

[0070] 104. Safe path generation and optimization: When the judgment result is not satisfied, based on the fusion state information, task target information and the current safety constraint rules, generate a motion path that satisfies the current safety constraint rules, and select one of them as a new motion path.

[0071] In this embodiment, a motion tree is constructed with the fused state information as the root node. The motion tree nodes store the state, depth, and control input. During the motion tree expansion stage, a sampling strategy based on model predictive control is invoked. The sampling strategy iteratively generates candidate control sequences within the prediction window.

[0072] In another possible implementation, the motion tree can be replaced by a continuous trajectory field based on sparse Gaussian process regression. The candidate control sequence is output as a complete trajectory by a deep reinforcement learning policy network (such as PPO) to obtain a set of feasible trajectories. The complete trajectory is forward simulated by a differentiable rigid-soft body coupled physics engine, and safety constraints are embedded into the reward through a barrier function method. Subsequently, a trust region policy optimization algorithm is used to select a trajectory from the set of feasible trajectories as a new motion path.

[0073] 105. New Path Execution Control: Control the cavity robot to execute the new motion path.

[0074] In this embodiment, the new motion path is parsed into a velocity command sequence; the velocity command sequence is input into an impedance control loop, which modifies the stiffness matrix according to real-time torque feedback to obtain a corrected stiffness matrix; the corrected stiffness matrix is ​​projected onto a Lie group to obtain an executable control quantity on the constrained manifold; the executable control quantity is transmitted to a servo driver to drive the cavity robot to perform motion.

[0075] In another possible implementation, the velocity command sequence can be processed by an adaptive model predictive control framework, and the constraint set can be adjusted online using a tubular robust controller based on Lyapunov functions. The stiffness matrix is ​​generated in real time through a hypernetwork obtained by reinforcement element learning, and then structure-preserving discretization is performed on the constraint manifold using a symplectic geometric integrator to obtain a low-drift control quantity. Finally, the control quantity is distributed to multi-core servo nodes for execution via an EtherCAT bus.

[0076] In some embodiments, the functions or modules of the apparatus provided in this application can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0077] The methods of the embodiments of this application have been described in detail above, and the apparatus of the embodiments of this application is provided below.

[0078] Please see Figure 2 , Figure 2 This is a schematic diagram of a real-time control device for a cavity robot based on dynamic safety boundaries, as proposed in an embodiment of this application. The real-time control device 1 includes: a safety boundary storage unit 11, a perception and rule triggering unit 12, a rule discrimination unit 13, a motion replanning unit 14, and a replanning execution unit 15. Specifically:

[0079] Security boundary storage unit 11: used to establish a dynamic security rule base, which stores security constraint rules defined based on linear time-series logic; the security constraint rules can be adjusted by changes in organizational characteristics sensed by operating instructions or sensors;

[0080] Perception and rule triggering unit 12: used to acquire sensor data of cavity robot, perform fusion processing on the sensor data to obtain fusion state information; the dynamic safety rule base activates the safety constraint rules of the corresponding subset as the current safety constraint rules according to the fusion state information;

[0081] Rule discrimination unit 13: used to determine whether the motion of the cavity robot satisfies the current safety constraint rule based on the fusion state information and the current safety constraint rule;

[0082] Motion replanning unit 14: When the judgment result is not satisfied, it generates a motion path that satisfies the current safety constraint rules based on the fused state information, task target information and the current safety constraint rules, and selects one of them as a new motion path;

[0083] Replanning execution unit 15: used to control the cavity robot to execute the new motion path.

[0084] In some embodiments, the functions or modules of the apparatus provided in this application can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0085] Please see Figure 3 , Figure 3A schematic diagram of the hardware architecture of an electronic device according to an embodiment of this application is shown. The electronic device 2 mainly consists of a processor 21 and a memory 22. In addition, the device may also include an input device 23 and an output device 24. The processor 21, memory 22, input device 23, and output device 24 are interconnected via connecting components, which can be various interfaces, data lines, or communication buses, etc., and are not specifically specified in this embodiment.

[0086] Processor 21 may be one or more graphics processing units (GPUs). If processor 21 is a GPU, the GPU may be single-core or multi-core. Alternatively, processor 21 may also be a processor group consisting of multiple GPUs interconnected via one or more buses. Furthermore, the processor may be other types of processors, which are not specifically limited in this embodiment.

[0087] The memory 22 is designed to store the instructions of a computer program and various program codes required to execute the present application. Optionally, the memory may include, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or optical disc read-only memory (CD-ROM), which are used to store related instructions and data.

[0088] Input device 23 is used to input data and / or signals, and output device 24 is used to output data and / or signals. Input device 23 and output device 24 can be independent devices or an integrated device.

[0089] It should be understood that, in the embodiments of this application, the memory 22 can store not only related instructions but also related data. The embodiments of this application do not specify the specific data content stored in the memory.

[0090] It should be understood that Figure 3 This illustration only shows a simplified design of an electronic device. In actual use, the electronic device may also include other necessary components, such as different numbers of input / output devices, processors, memory, etc. All electronic devices capable of implementing the embodiments of this application are within the protection scope of this application.

[0091] Those skilled in the art will recognize that the components and algorithm steps of the various examples described in the embodiments disclosed herein can be implemented by electronic hardware or by a combination of computer software and electronic hardware. Whether these functions are implemented through hardware or software will be determined based on the specific application requirements and design constraints of the technical solution. Those skilled in the art can adopt different implementation methods according to the needs of each specific application, but such implementation should not be considered as exceeding the scope of protection of this application.

[0092] Those skilled in the art should understand that, for ease of description and simplification, the specific operational procedures of the aforementioned systems, devices, and components can be referred to the corresponding steps in the preceding method embodiments, and will not be repeated here. Furthermore, those skilled in the art should also understand that each embodiment in this application has its own focus, and for ease of description and simplification, the same or similar content may not be repeated in different embodiments. Therefore, if a part is not mentioned or not described in detail in a certain embodiment, it can be referred to the relevant description in other embodiments.

[0093] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can also be implemented through other means. For example, the described apparatus embodiments are merely exemplary, and the division of the units therein is only a logical functional division; different division methods may exist in actual implementation. For example, multiple units or components may be merged or integrated into another system, or certain features may be omitted, or certain steps may not be performed. Furthermore, the interconnections shown or discussed, whether direct or indirect, whether coupling or communication connections, may be implemented electrically, mechanically, or otherwise through interfaces, devices, or units.

[0094] Units described as independent components may or may not actually be physically separate; parts presented as units may or may not be physical entities, meaning they may be concentrated in one location or distributed across multiple network nodes. Depending on actual needs, some or all of these units can be selected to achieve the objectives of this embodiment.

[0095] Furthermore, in the various embodiments of this application, each functional unit can be integrated into a single processing unit, exist independently, or two or more units can be merged into one unit. In the foregoing embodiments, the relevant functions can be fully or partially implemented through software, hardware, firmware, or any combination thereof. If software implementation is chosen, it can be implemented entirely or partially in the form of a computer program product. This computer program product contains one or more computer instructions. When these instructions are loaded and executed on a computer, they will produce all or part of the processes or functions described in the embodiments of this application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. These computer instructions can be stored in computer-readable storage media or transmitted through such media. Computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, DSL) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any computer-accessible available medium, or a data storage facility such as a server or data center that integrates one or more available media. These available media may include magnetic media (such as floppy disks, hard disks, and magnetic tapes), optical media (such as DVDs), semiconductor media (such as SSDs), etc. Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by computer program instructions and related hardware, and these programs can be stored in computer-readable storage media. When these programs are executed, they will contain the processes of the above method embodiments. The aforementioned storage media include, but are not limited to, various media capable of storing program code, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A real-time control method for a cavity robot based on dynamic safety boundaries, characterized in that, The method comprises the following steps: S1: establishing a dynamic safety rule base, wherein the dynamic safety rule base stores safety constraint rules defined based on linear temporal logic; the safety constraint rules allow adjustment of changes in tissue characteristics perceived by operation instructions or sensors; specifically comprising: constructing a set of first-order linear temporal logic formulas, wherein the formula set takes luminal tissue elastic modulus, wall thickness, temperature and contact force as atomic propositions; using a runtime verification framework to compile the formula set into a finite state monitor, wherein the finite state monitor takes event stream as input and outputs Boolean label; registering the finite state monitor to a rule update interface, wherein the rule update interface triggers recompilation of the finite state monitor when receiving tissue characteristic change events; the recompilation includes calling a syntax rewriting engine, which inserts or deletes atomic propositions while maintaining semantic equivalence of the formula set; S2: obtaining sensor data of the luminal robot, fusing the sensor data to obtain fused state information; the dynamic safety rule base activates the safety constraint rules of the corresponding subset as the current safety constraint rules according to the fused state information; specifically comprising: synchronously collecting intraluminal images, fiber Bragg grating strain, electromagnetic tracking pose and miniature ultrasonic echoes through a multi-modal sensor array; inputting the intraluminal images into a feature encoding network based on self-supervised learning to obtain a dense geometric feature tensor; inputting the fiber Bragg grating strain, the electromagnetic tracking pose and the miniature ultrasonic echoes into a cross-domain alignment module based on a graph neural network to output a state node vector in a unified coordinate system; performing extended Kalman consistency filtering on the state node vector to obtain the fused state information; the dynamic safety rule base listens to the characteristic change field in the fused state information and indexes and matches the subset rules according to the field value; S3: judging whether the motion of the luminal robot satisfies the current safety constraint rules based on the fused state information and the current safety constraint rules; S4: when the judgment result is not satisfied, generating a motion path that satisfies the current safety constraint rules based on the fused state information, task target information and the current safety constraint rules, and selecting one of them as a new motion path; S5: controlling the luminal robot to execute the new motion path.

2. The method of claim 1, wherein, Step S3 comprises: transforming the fused state information into a symbolic trajectory indexed by discrete time steps; transforming the current safety constraint rules into a timed automaton; performing online model checking of the timed automaton on the symbolic trajectory to output a binary label of satisfaction or dissatisfaction; when the label is not satisfied, recording the first time step and the corresponding state variable where the violation occurs.

3. The method of claim 1, wherein, Step S4 comprises: constructing a motion tree with the fused state information as the root node, wherein the motion tree nodes store state, depth and control input; in the motion tree expansion phase, calling a sampling strategy based on model predictive control, which iteratively generates candidate control sequences within a prediction window; inputting the candidate control sequences into a physics engine, which outputs a predicted trajectory; verify the predicted trajectory point by point using the current safety constraint rule, and retain the trajectories passing the verification; run a path planning algorithm on the retained trajectory set to obtain candidate paths and path scores; select the candidate path with the highest score as the new motion path.

4. The method of claim 1, wherein, Step S5 includes: parse the new motion path into a velocity instruction sequence; input the velocity instruction sequence into an impedance control loop, which modifies the stiffness matrix according to real-time torque feedback to obtain a corrected stiffness matrix; perform Lie group projection on the corrected stiffness matrix to obtain executable control quantities on the constraint manifold; pass the executable control quantities to a servo driver to drive the lumen robot to execute motion.

5. A dynamic security perimeter based real-time control device for a lumenal robot, comprising: comprise: a safety boundary storage unit: establish a dynamic safety rule library, which stores safety constraint rules defined based on linear temporal logic; the safety constraint rules allow adjustment of changes in tissue characteristics perceived by the operation instructions or sensors; specifically including: construct a set of first-order linear temporal logic formulas, the formula set takes the lumen tissue elastic modulus, wall thickness, temperature and contact force as atomic propositions; use a runtime verification framework to compile the formula set into a finite state monitor, which takes an event stream as input and outputs a Boolean flag; register the finite state monitor to a rule update interface, which triggers the finite state monitor to recompile when receiving a tissue characteristic change event; the recompilation includes calling a syntax rewriting engine, which inserts or deletes atomic propositions while maintaining the semantic equivalence of the formula set; a perception and rule triggering unit: obtain sensor data of the lumen robot, fuse the sensor data to obtain fused state information; the dynamic safety rule library activates the safety constraint rules in the corresponding subset as the current safety constraint rules according to the fused state information; specifically including: synchronously collect intraluminal images, fiber Bragg grating strain, electromagnetic tracking pose and miniature ultrasonic echoes through a multi-modal sensor array; input the intraluminal images into a feature encoding network based on self-supervised learning to obtain a dense geometric feature tensor; input the fiber Bragg grating strain, electromagnetic tracking pose and miniature ultrasonic echoes into a cross-domain alignment module based on graph neural networks to output a state node vector in a unified coordinate system; perform extended Kalman consistency filtering on the state node vector to obtain the fused state information; the dynamic safety rule library listens to the characteristic change field in the fused state information and indexes and matches the subset rules according to the field value; a rule discrimination unit: used to determine whether the motion of the lumen robot satisfies the current safety constraint rule based on the fused state information and the current safety constraint rule; a motion re-planning unit: used to generate a motion path that satisfies the current safety constraint rule based on the fused state information, task target information and the current safety constraint rule when the determination result is not satisfied, and select one as a new motion path; A re-planning execution unit for controlling the lumen robot to execute the new motion path.

6. An electronic device, comprising: Comprising: A processor and a storage unit for storing computer program code, said code containing computer instructions, when executed by the processor, cause the electronic device to perform the method of any one of claims 1 to 4.

7. A computer readable storage medium characterized in that, A computer readable storage medium having stored therein a computer program, said computer program containing program instructions, when executed by a processor, cause the processor to perform the method of any one of claims 1 to 4.

8. A computer program product, characterised in that, The computer program product contains computer programs or instructions, when the computer programs or instructions run on the computer, make the computer execute the method of any one of claims 1 to 4.

Images