An autonomous navigation control system for a reconfigurable ferrofluid droplet robot

Abstract

An autonomous navigation control system for a reconfigurable ferrofluid droplet robot includes: a distributed magnetic field generation unit, which is composed of multiple electromagnets and can generate a local non-uniform gradient magnetic field; an environmental perception unit uses a visual system to detect the position information of the robot and obstacles in real time, and performs multi-target detection, tracking and positioning; a motion planning unit uses a path planning algorithm based on environmental information to plan a collision-free path for the robot formation and update it in real time; a closed-loop controller adjusts the robot's motion state based on feedback information and dynamically controls its shape change to adapt to environmental changes; a drive control unit controls the coordinated movement of the robot group based on control signals. The system design of the present invention solves the problems of poor adaptability and single functional form of traditional micro-nano robot systems, significantly improves the flexibility and efficiency of operations, and broadens its application prospects in the fields of drug delivery, biosensing, environmental remediation, etc.

Description

Technical Field

[0001] The present invention relates to the field of micro-nano robots suitable for medical surgery, and in particular to an autonomous navigation control system of a reconfigurable ferromagnetic fluid droplet robot. Background Art

[0002] Magnetic field-driven micro-nano robots have the significant advantages of being small and remotely operable. By regulating the external magnetic field in time and space, micro-nano robots can be precisely controlled, and have broad application prospects in drug delivery, biosensing, environmental remediation, and active material assembly. However, limited by closed-loop control systems, micro-nano robots can only move toward a specific target point along a predetermined path in two-dimensional or three-dimensional space, and cannot adapt to complex and changing task requirements; in addition, due to the nature of the interaction between magnetic materials and the global magnetic field, the independent control and collaborative planning of robot groups are full of challenges. In particular, the automated formation collaboration of reconfigurable droplet robots with adaptive morphological transformation capabilities can bring the challenges of manipulation tasks to a new level.

[0003] It should be noted that the information disclosed in the above background technology section is only used for understanding the background of the present application, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the invention

[0004] The main purpose of the present invention is to overcome the defects existing in the above-mentioned background technology and provide an autonomous navigation control system of a reconfigurable ferrofluid droplet robot.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] An autonomous navigation control system of a reconfigurable ferrofluid droplet robot, comprising:

[0007] Multiple ferrofluid droplet robots, with superparamagnetic and liquid properties, can generate induced magnetic moments under the action of an external magnetic field, align with the direction of the external magnetic field, and can change shape to minimize internal energy. They can be subjected to a magnetic force that is positively correlated to their volume size under a magnetic field gradient; they have the ability to be reconfigured programmably, including separation, merging, and shape change;

[0008] A distributed magnetic field generating unit, which includes a plurality of electromagnets arranged in an array, and is used to generate a local non-uniform gradient magnetic field, so as to realize addressable operation and group cooperative driving of the ferrofluid droplet robot;

[0009] The environmental perception unit detects the real-time position information of the ferrofluid droplet robot and obstacles through the visual system, and performs multi-target position detection, tracking and positioning of the ferrofluid droplet robot group and environmental information perception;

[0010] The motion planning unit detects and handles path conflicts based on the position and obstacle information of each ferrofluid droplet robot through a path planning algorithm, and plans a collision-free motion path for the ferrofluid droplet robot formation;

[0011] A closed-loop controller adjusts the motion state of the ferrofluid droplet robot according to feedback information from the environment sensing unit, so that it moves along the path planned by the motion planning unit to adapt to different motion requirements, and dynamically controls the shape change of the droplet robot to adapt to changes in the surrounding environment to achieve closed-loop feedback control;

[0012] The driving control unit is connected to the closed-loop controller and drives the electromagnets in the distributed magnetic field generating unit according to the control signal thereof, so as to precisely control the coordinated movement of the ferrofluid droplet robot group.

[0013] Furthermore, the driving control unit includes a communicating and interactive computer and an embedded system. The computer serves as the central processing unit of the system, processes control signals and generates control instructions. The embedded system receives instructions from the computer, controls the current driving board to output current, so as to modulate the distributed local magnetic field and realize programmable movement and deformation of the droplet robot.

[0014] Furthermore, the distributed magnetic field generating unit includes a plurality of electromagnets arranged in an array, each electromagnet is individually connected to a power source, and when currents of different magnitudes, directions and frequencies are passed through, a specific time-varying magnetic field can be generated as required. The magnitude and range of the magnetic field are related to the applied current. The magnetic field generated by a single electromagnet is expressed as:

[0015]

[0016] in µ 0 is the vacuum permeability, N is the total number of turns of the coil, I is the current passing through the coil, M is the magnetization intensity of the iron core, z is the distance from the electromagnet surface along the z-axis in space, L is the length of the coil winding, R is the radius of the coil, and r are the length and radius of the core respectively;

[0017] When the ferrofluid droplet robot is in the non-uniform gradient field generated by the electromagnet, the ferrofluid will be magnetized and thus affected by the external magnetic field. The magnetic force is expressed as:

[0018]

[0019] in V represents the volume of the magnetic fluid droplet, m represents the magnetization intensity of the magnetic fluid droplet, and B represents the magnetic flux density of the external magnetic field.

[0020] Further, the driving mode of the distributed magnetic field generating unit includes a single-coil driving mode and a double-coil driving mode;

[0021] In the single-coil driving mode, by activating the coil adjacent to the ferrofluid droplet robot, a local magnetic field is generated to magnetize the paramagnetic droplet, thereby attracting the ferrofluid droplet robot to move to the center of the coil;

[0022] In the dual-coil driving mode, by applying opposite currents to adjacent coils, the ferrofluid droplet robot is first stretched into a strip under the action of the magnetic field, and then the ferrofluid droplet robot is prompted to maintain streamlined movement by changing the activated coil. At the same time, by applying the same current to adjacent coils, the ferrofluid droplet robot is induced to split, so that it can pass through a narrow confined space smaller than itself.

[0023] Furthermore, the environmental perception unit detects static and dynamic objects in the workspace, including obstacle positions, unstructured scene features, and motion parameters of ferrofluid droplet robots, and performs scene recognition, map construction, and multi-target tracking and positioning, specifically including: acquiring the first frame image of the camera, using a machine vision algorithm based on a support vector machine to identify and classify obstacles, feasible domains, and ferrofluid droplet robots, extracting potential path points in the workspace, boundary information of obstacles, and the number of droplet robots and the corresponding coordinates of each robot; mapping the real map into the simulation environment to construct a map model; tracking the robot through a multi-target tracking algorithm and updating its position coordinates in real time, and calibrating the position of the electromagnet to activate the corresponding coil to independently drive the droplet robot.

[0024] Furthermore, based on the perception of environmental information, the ferrofluid droplet robot is controlled to adapt to the unstructured environment by deforming or splitting into smaller sub-droplets during movement; the movement process is divided into three states: an initial state, in which the ferrofluid droplet robot moves controllably under the drive of a single coil; a stretching state, in which the ferrofluid droplet robot deforms into a long strip and moves under the drive of multiple coils; and a splitting state, in which the ferrofluid droplet robot splits into multiple sub-droplets and moves.

[0025] Furthermore, the environment perception unit performs shape detection on the robot during the robot's form switching process, including:

[0026] The image preprocessing subunit is used to convert the color image acquired by the camera into a grayscale image, perform a Gaussian blur operation to smooth the image, and then use the Canny edge detection algorithm to obtain the edge information of the robot in the image;

[0027] A feature extraction subunit is used to match the target shape with a known template using a template matching-based method, find similarities, and classify using the geometric features of the shape;

[0028] The shape recognition subunit is used to recognize the shape of the ferrofluid droplet robot according to the extracted geometric features.

[0029] Furthermore, the closed-loop controller adopts a finite state machine to construct a mutual conversion strategy between various modes of the reconfigurable droplet robot, wherein a corresponding state table and action query table are established, and the state is switched according to the trigger event; wherein, when there is no obstruction on the front path, the ferrofluid droplet robot is controlled to maintain the initial state movement; when there is a confined space on the front path, the ferrofluid droplet robot is controlled to deform into a long strip movement or split into sub-droplet movement according to the spatial characteristics.

[0030] Furthermore, the motion planning unit uses incremental conflict search (ICBS) path planning for path planning in a dynamically changing working environment, specifically including:

[0031] A conflict information retention unit is used to retain previous search information, especially path conflict information, when the environment or task changes;

[0032] The initial path planning subunit is used to calculate the initial paths for all robots in the initial situation and detect conflicts, dividing the problem into high-level conflict detection and low-level single robot path search;

[0033] Dynamic change detection unit, used to detect changes in the environment or task, including new obstacles, changes in the number of robots, and changes in the target point, and identify the affected robots;

[0034] Incremental conflict handling unit, used to make local adjustments to the affected robot paths, minimize the conflict area, reduce the cost of path planning and calculation, and ensure that the changed planning results are still conflict-free;

[0035] Among them, the high-level conflict detection creates constraints and divides the conflict into multiple branches when a multi-robot path conflict is detected, while the low-level path search uses a classic path planning algorithm (such as D*) to solve the path for each robot while satisfying the constraints passed from the high-level until the paths of all robots are free of conflict.

[0036] Furthermore, the motion planning unit further comprises:

[0037] The static obstacle avoidance planning subunit is used to plan a collision-free path for the robot group when only static obstacles exist, so that the robots meet the physical constraints and avoid obstacles to reach the target point;

[0038] The conflict tree construction subunit is used to construct a conflict tree at a high level when a robot path conflict is detected, and generate two child nodes for each conflict in the conflict tree, each of which constrains the path of a conflicting robot to force avoidance of the conflict;

[0039] The path update subunit is used to update the path using the D* path planning algorithm at the low level and continue to detect new conflicts after the update until all conflicts are resolved;

[0040] The environment change response subunit is used to receive the change information transmitted by the environment perception unit when new obstacles appear in the workspace, and make local path adjustments according to the new environment map to achieve obstacle avoidance;

[0041] The dynamic path adjustment subunit is used to re-plan the feasible path according to the updated environment map when multiple new obstacles are detected in the environment, so that the robot can successfully reach the target location;

[0042] The static obstacle avoidance planning subunit ensures that the distance constraint between the robots is set to the length of two coils to avoid mutual interference of the magnetic fields of adjacent coils, and ensures that there is no conflict between the robots in the same time step when there is a path intersection.

[0043] The present invention has the following beneficial effects:

[0044] The present invention provides an autonomous navigation and control system for a reconfigurable ferrofluid droplet robot, which realizes the fully automated operation of a group of droplet robots. The present invention innovatively integrates the robot's environmental perception, motion planning, modal conversion and drive control. The motion planning unit calculates the optimal path for the reconfigurable robot formation based on the perceived environmental information, and dynamically updates it in real time; the drive control unit drives the droplet robot to move toward the target point according to the planned path, and performs modal conversion according to the environmental information during the movement. The above design can solve the problems of poor adaptability and single functional form faced by traditional micro-robot systems, and provides a solution for the autonomous navigation control of reconfigurable ferrofluid droplet robots.

[0045] The present invention provides an efficient and flexible operation scheme for a group of reconfigurable ferrofluid droplet robots, breaking through the limitations of traditional micro-nano robots in terms of adaptability and functional diversity. The system adopts a distributed magnetic field generation platform, combined with array-arranged electromagnets, to achieve independent addressable operation and group collaborative drive of droplet robots, enabling the robots to navigate autonomously in complex environments. Through a vision-guided method, the system can achieve adaptive morphological transformation of robots in restricted scenarios, while the multi-robot motion planning strategy ensures effective collaboration of the formation in the face of dynamic changes. The closed-loop control system uses real-time feedback from the environmental perception unit to dynamically adjust the motion state and morphology of the droplet robot to adapt to the changing task requirements. This integrated design not only improves the accuracy and adaptability of the operation, but also expands the application potential of droplet robots in the fields of drug delivery, biosensing, environmental remediation, and active material assembly, showing a high degree of innovation and practicality.

[0046] Other beneficial effects of the embodiments of the present invention will be further described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 Schematic diagram of an autonomous navigation control system according to an embodiment of the present invention.

[0048] Figure 2 Schematic diagram of the mode conversion of the ferrofluid droplet robot according to an embodiment of the present invention.

[0049] Figure 3 Schematic diagram of static obstacle avoidance of the ferrofluid droplet robot according to an embodiment of the present invention.

[0050] Figure 4 Schematic diagram of dynamic obstacle avoidance of a ferrofluid droplet robot according to an embodiment of the present invention.

[0051] Figure 5 Schematic diagram of autonomous navigation of a ferrofluid droplet robot according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The following is a detailed description of the embodiments of the present invention. It should be emphasized that the following description is only exemplary and is not intended to limit the scope and application of the present invention.

[0053] It should be noted that when an element is referred to as being "fixed to" or "disposed on" another element, it can be directly on the other element or indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or indirectly connected to the other element. In addition, connection can be used for fixing as well as for coupling or communication.

[0054] It should be understood that the orientation or position relationship indicated by terms such as "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0055] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0056] In response to the above problems and technical requirements, the present invention proposes an autonomous navigation control system for a group of reconfigurable ferrofluid droplet robots. A distributed local magnetic field generation platform is designed to achieve independent addressable operations of multiple droplet robots; combined with a method based on visual guidance, the adaptive morphological transformation of robots in restricted scenes is achieved; a multi-robot motion planning strategy is constructed to achieve autonomous navigation of a group of reconfigurable ferrofluid droplet robots in complex environments.

[0057] See also Figures 1 to 5The embodiment of the present invention provides an autonomous navigation control system of a reconfigurable ferrofluid droplet robot, comprising: a plurality of ferrofluid droplet robots, which have superparamagnetism and liquid properties, can generate induced magnetic moments under the action of an external magnetic field, realize alignment with the direction of the external magnetic field, can change shape to minimize internal energy, and can be subjected to a magnetic force positively correlated with its volume size under a magnetic field gradient; the ferrofluid droplet robot has programmable reconfiguration capabilities, including separation, merging and shape change; a distributed magnetic field generation unit, which includes a plurality of electromagnets arranged in an array, for generating a local non-uniform gradient magnetic field, realizing addressable operation and group collaborative driving of the ferrofluid droplet robot, and the distributed magnetic field generation unit is equipped with a power module for power supply; an environmental perception unit (such as a CCD camera), which detects the ferrofluid droplet robot and obstacles through a visual system Real-time position information is used to perform multi-target position detection, tracking and positioning of the ferrofluid droplet robot group and environmental information perception; the motion planning unit detects and handles path conflicts based on the position and obstacle information of each ferrofluid droplet robot through a path planning algorithm, and plans a collision-free motion path for the ferrofluid droplet robot formation; a closed-loop controller adjusts the motion state of the ferrofluid droplet robot according to the feedback information from the environmental perception unit, so that it moves along the path planned by the motion planning unit to adapt to different motion requirements, and dynamically controls the shape change of the droplet robot to adapt to changes in the surrounding environment, so as to achieve closed-loop feedback control; a drive control unit is connected to the closed-loop controller, and drives the electromagnet in the distributed magnetic field generation unit according to its control signal, so as to accurately control the coordinated motion of the ferrofluid droplet robot group.

[0058] In some embodiments, the ferromagnetic fluid may include a surfactant, a carrier liquid and dispersed magnetic nanoparticles Fe3O4 to form a colloidal suspension. The present invention does not limit the specific composition of the ferromagnetic fluid.

[0059] See also Figures 2 to 5 In some embodiments, the plurality of electromagnets are regularly arranged to form a square array.

[0060] In some embodiments, an autonomous navigation and control system for a swarm of reconfigurable ferrofluid droplet robots includes an electromagnet array design for driving the droplet robots, a visual detection method for complex environment perception, a finite state machine construction for robot modal conversion, and the formulation of collaborative motion planning strategies for multiple robots.

[0061] The fully automatic navigation system includes a robot environment perception unit, a robot motion planning unit and a robot drive control unit. The robot environment perception unit provides real-time positioning for the magnetic field-driven droplet robot, and dynamically adjusts the shape of the reconfigurable droplet robot by detecting the surrounding environment; the robot motion planning unit is used to plan an ideal collision-free driving path for the robot formation according to the position and obstacle information of each robot; the robot drive control unit includes a communication and interactive computer and embedded system, which is used to receive control signals, output rated current, generate the desired time-varying magnetic field, and realize the programmable movement and deformation of the droplet robot.

[0062] The robot drive control unit is used to receive input signals and modulate the distributed local magnetic field. In a specific embodiment, the distributed magnetic field is generated by 144 electromagnets arranged in a 12×12 array. Each electromagnet is individually connected to a power supply, and when currents of different magnitudes, directions, and frequencies are passed through it, it can generate a specific time-varying magnetic field on demand. The magnitude and range of the magnetic field are related to the applied current. The magnetic field generated by a single electromagnet can be expressed as:

[0063]

[0064] in µ 0 is the vacuum permeability, N is the total number of turns of the coil, I is the current passing through the coil, M is the magnetization intensity of the iron core, z is the distance from the electromagnet surface along the z-axis in space, L is the length of the coil winding, R is the radius of the coil, and r are the length and radius of the iron core respectively. When the droplet robot is in the non-uniform gradient field generated by the electromagnet, the ferromagnetic fluid will be magnetized and thus affected by the external magnetic field. The magnetic force can be expressed as:

[0065]

[0066] in V represents the volume of the magnetic fluid droplet, m represents the magnetization intensity of the magnetic fluid droplet, and B represents the magnetic flux density of the external magnetic field.

[0067] The droplet robot has both liquid properties and magnetism, and can produce extreme deformation under the action of an external magnetic field to adapt to unstructured environments. In the single-coil drive mode, by activating the adjacent coils of the droplet robot, a local magnetic field is generated to magnetize the paramagnetic droplets, attracting the droplet robot to move to the center of the coil; in the dual-coil drive mode, by applying opposite currents to the adjacent coils, the droplet robot is first stretched into a strip under the action of the magnetic field, and then the activated coil is changed to cause the droplet robot to maintain streamlined movement. At the same time, by applying the same current to the adjacent coils, the dual coils can also induce the robot to split, so that it can pass through narrow confined spaces that are much smaller than itself.

[0068] The robot environment perception unit is used to detect static and dynamic objects in the workspace, including obstacle positions, unstructured scene features, and motion parameters of the droplet robot, providing visual feedback for the system's automated closed-loop control. It mainly includes: scene recognition, map construction, and multi-target tracking and positioning. The specific processing is as follows: First, the first frame of the camera is obtained, and the machine vision algorithm based on support vector machine is used to identify and classify obstacles, feasible domains, and droplet robots, extract potential path points in the workspace, boundary information of obstacles, and the number of droplet robots and the corresponding coordinates of each robot, and further map the real map to the simulation environment to build a map model. On this basis, a suitable OpenCV multi-target tracking algorithm is selected to track the robot and update its position coordinates in real time, and the position of the electromagnet is calibrated to activate the corresponding coil to independently drive the droplet robot.

[0069] The robot can adapt to unstructured environments by deforming and splitting into smaller sub-droplets during movement. The whole process is divided into three states: initial state movement, that is, the robot moves in a controllable manner under the drive of a single coil. This state is convenient and efficient, and can maximize the number of droplet robots that can exist in the workspace at the same time; stretching state movement, that is, the robot deforms into a long strip under the drive of multiple coils, and this state allows the robot to pass through a relatively narrow space; split state movement, that is, the robot splits into multiple sub-droplets, which allows the robot to greatly reduce its own size and pass through a channel much smaller than itself.

[0070] The finite state machine is used to construct the mutual conversion strategy between various modes of the reconfigurable droplet robot. The corresponding state table and action query table are established to switch the state according to the trigger event. For example, when there is no obstacle in the path ahead of the robot, the robot maintains the initial state movement; when there is a confined space in the path ahead of the robot, the robot chooses to deform into a long strip movement or split into sub-droplet movement according to the spatial characteristics.

[0071] The robot motion planning unit plans a collision-free feasible path for the droplet robot formation to achieve autonomous navigation of the droplet robot formation. In order to effectively complete dynamic and static obstacle avoidance, an incremental conflict search (ICBS) path planning algorithm is proposed. When the environment or task changes, the algorithm retains the previous search information, especially the conflict information of the path, and performs incremental updates on this basis instead of starting a new plan from scratch.

[0072] The specific process of the incremental conflict search (ICBS) path planning algorithm is as follows: 1. Initial path planning: In the initial case, the ICBS algorithm is the same as the classic CBS, calculating the initial path for all robots and detecting conflicts. This method divides the problem into high-level conflict detection and low-level single robot path search. At the high level, the path conflicts of multiple robots are detected each time. When conflicts are found, constraints are created and the conflicts are divided into multiple branches. At the low level, the classic path planning algorithm (such as D*) is used to solve the path for each robot, while satisfying the constraints passed from the high level. When there is no conflict in the paths of all robots, the algorithm stops. 2. Detection of dynamic changes: When the environment or task changes, such as the appearance of new obstacles, changes in the number of robots, changes in the target point, etc. At this time, the affected paths are detected by the environmental perception unit and the affected robots are identified. 3. Incremental conflict handling: For the affected robot paths, ICBS minimizes the conflict area by locally adjusting the path, thereby reducing the cost of path planning and calculation, and ensuring that the changed planning results are still conflict-free.

[0073] Examples

[0074] like Figure 1As shown in the figure, an autonomous navigation control system of a reconfigurable ferrofluid droplet robot includes a motion planning unit, a drive control unit, a distributed magnetic field generation unit and an environment perception unit. The motion planning unit is used to generate the desired path for the droplet robot. According to the initial position and target position of each robot, an improved incremental conflict search path planning algorithm is used to plan a collision-free feasible path for the robot formation. At the same time, the planned path is dynamically updated according to the environmental characteristics to realize the adaptive navigation of the reconfigurable droplet robot. The algorithm is written in Python. Since the robot is a discrete point-to-point motion, the grid map method is used to construct the environmental model. When the number of robots, obstacle positions or target points in the workspace change, the incremental conflict search algorithm can retain the previous search information, locally adjust the conflict area, and realize path planning in a dynamic environment. The drive control unit is composed of a computer and an embedded system. The computer sends control instructions to the single-chip microcomputer through the Modbus protocol, and a PID closed-loop controller is designed to realize the position control, speed control and shape control of the droplet robot. The embedded system includes a single-chip microcomputer and a customized current drive board, which communicate with each other through the IIC bus. Each current drive board has a custom preset address. Only two signal lines are needed to control the entire platform. The number of current control boards can be selected according to the size of the electromagnetic coil array. Theoretically, up to 576 electromagnets can be independently controlled. The distributed magnetic field generation unit can generate the desired local magnetic field on demand to drive the droplet robot to move or produce extreme deformation. The magnetic field range of each coil only includes the four adjacent upper, lower, left and right coils. Only when the droplet robot is within the magnetic field range of the activation coil can it be effectively driven. This discrete local magnetic field does not require a complex decoupling process to achieve independent drive and collaborative control of the robot formation. The environmental perception unit uses a CCD camera to record the environmental information of the workspace at 50 frames per second, providing the real-time position and environmental characteristics of the robot, and providing visual feedback for the motion planning and closed-loop control of the robot system. The robot needs to be subjected to shape detection during the morphology switching process, which includes three steps: image preprocessing, feature extraction, and shape recognition. Image preprocessing is to reduce noise and improve image quality. First, the color image acquired by the camera is converted into a grayscale image to reduce computational complexity. Then, a Gaussian blur operation is performed to smooth the image and reduce the impact of noise. Then, the Canny edge detection algorithm is used to obtain the edge information of the robot in the image. After image preprocessing, the features of the target shape need to be extracted. A template matching-based method is used to match the target shape with a known template to find similarities. After feature extraction, the geometric features of the shape (such as area, perimeter, and angle) are used to classify it, thereby identifying the morphology of the reconfigurable droplet robot.

[0075] like Figure 2As shown in Figure 1, the magnetic fluid droplet robot has various forms and can be transformed into different states through the local magnetic field generated by the electromagnet array. Figure 2 As shown in (a) in the figure, when a single coil is in action, the paramagnetic droplet is first magnetized under the action of the external magnetic field, and then attracted to the center of the coil under the combined influence of the magnetic force and surface tension to maintain a stable state. Figure 2 As shown in (b) and (c) in the figure, when opposite currents are passed through adjacent coils, the magnetic fluid droplet will be elongated. The more coils are activated at the same time, the greater the deformation of the droplet robot. This morphology allows the robot to pass through confined narrow spaces. Figure 2 As shown in (d), when three adjacent coils apply current in the same direction, the droplet robot will split into multiple sub-droplets because the magnetic force is greater than the surface tension. This can greatly reduce the volume of the droplet, allowing it to pass through some particularly narrow channels. The reconfigurable ability of the droplet robot enables it to adapt to complex and changing environments. A finite state machine is designed to achieve automatic switching of the robot's form. The conditions for state transfer are: when there is no confined space in the robot's forward direction, the robot adopts a single-coil drive mode, that is, it maintains the initial state movement; when there is a confined space in the robot's forward direction, the robot adopts a multi-coil drive mode, and chooses to stretch or split into sub-droplets according to the characteristics of the confined space.

[0076] like Figure 3 As shown in the figure, an improved incremental conflict search path planning algorithm is used to plan a collision-free path for the robot group. When there are only static obstacles, the algorithm is the same as the traditional conflict search algorithm. It can not only meet the physical constraints between robots but also avoid obstacles and reach the target point. The algorithm first plans an initial path for each robot; then, by simulating the path of the robot, it detects whether there is a conflict. For each conflict detected, two child nodes are generated in the high-level conflict tree, each of which constrains the path of a conflicting robot and forces it to avoid the conflict; after updating the path using the low-level D* path planning algorithm, new conflicts continue to be detected and decomposed at a high level until all conflicts are resolved. Take the case of three robots in the workspace, which are represented by circles, numbered 0, 1, and 2, and the corresponding target points are represented by squares with digital labels. Due to the influence of the magnetic field of adjacent coils, the distance constraint between the robots is set to the length of two coils. The lines with arrows represent the feasible paths of the three robots. Although there are path intersections, there is no conflict between the robots at the same time step.

[0077] like Figure 4As shown in the figure, the incremental conflict search algorithm uses the D* path planning method at the low level. The robot system can not only achieve static obstacle avoidance, but also adapt to the dynamically changing environment. When new obstacles suddenly appear in the workspace, the environmental perception unit can transmit the changes to the motion planning unit and respond in time. Figure 4 As shown in (a) in the figure, in the initial state, the algorithm plans feasible paths that meet the constraints for the two robots respectively. Figure 4 When the robot 0 and robot 1 are in position (b), obstacles O1 and O2 suddenly appear in their forward direction. At this time, the environment perception unit detects the change of obstacles and feeds the information back to the motion planning unit. The motion planning unit adjusts the local path according to the new environment map to achieve the purpose of obstacle avoidance. Figure 4 When the robot reaches position (c) in the figure, obstacles O3 and O4 are added to the environment. The motion planning unit will re-plan feasible paths based on the updated environment map, and finally the robots will successfully reach their respective target positions.

[0078] like Figure 5 As shown in the figure, the autonomous navigation control system can realize the adaptive 2D navigation of the reconfigurable magnetofluid droplet robot in a complex environment. First, the environmental perception unit obtains the information of the robot and obstacles in the workspace, and the motion planning unit plans a feasible path for the robot. When it detects that there is a confined space in the forward direction, the robot can be deformed into a long strip to pass through the confined space, and then return to the initial state to continue moving forward; when a narrow channel is detected, the robot can be further split into multiple sub-droplets to pass through the narrow channel. At this time, due to the increase in the number of robots, the motion planning unit will reallocate the path according to the number of robots detected and their respective target points. In addition, the distributed local magnetic field generated by the electromagnet array allows the droplet robot formation to collaborate in parallel, which can significantly improve the efficiency of task execution. After passing through the narrow channel, multiple sub-droplets will be merged according to task requirements. Combined with the proposed dynamic path planning algorithm, the robot formation can realize autonomous navigation in a dynamic environment and finally reach their respective designated positions.

[0079] In summary, the system of the present invention provides a fully automated operation scheme for the reconfigurable ferrofluid droplet robot group through an innovative integrated design scheme, which greatly enhances the adaptability and functionality of traditional micro-nano robots. The system cleverly integrates multiple key links such as environmental perception, motion planning, modal conversion and drive control, and can autonomously plan the optimal path based on real-time environmental information and dynamically update it. The design of the distributed magnetic field generation platform and the array-arranged electromagnets realizes the precise control and group collaborative drive of the droplet robot, making it possible for the robot to navigate autonomously in a complex environment. The system realizes the adaptive morphological transformation of the robot in a restricted scene through a vision-based method, and the closed-loop control system ensures that the robot can adjust the motion state and morphology in real time according to environmental changes to meet the changing task requirements. The system design of the present invention not only improves the accuracy and adaptability of the operation, but also broadens the application scope of the droplet robot in multiple fields, such as drug delivery, biosensing, environmental remediation and active material assembly, etc., reflecting a high degree of innovation and practicality.

[0080] The above content is a further detailed description of the present invention in combination with specific / preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, without departing from the concept of the present invention, it can also make several substitutions or modifications to these described embodiments, and these substitutions or modifications should be regarded as belonging to the protection scope of the present invention. In the description of this specification, the description of the reference terms "an embodiment", "some embodiments", "preferred embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in combination with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In the absence of mutual contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples. Although the embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions and modifications can be made herein without departing from the scope of protection of the patent application.

Claims

1. An autonomous navigation control system for a reconfigurable ferrofluid droplet robot, characterized in that: include: Multiple ferrofluid droplet robots, with superparamagnetic and liquid properties, can generate induced magnetic moments under the action of an external magnetic field, align with the direction of the external magnetic field, and can change shape to minimize internal energy. They can be subjected to a magnetic force that is positively correlated to their volume size under a magnetic field gradient; they have the ability to be reconfigured programmably, including separation, merging, and shape change; A distributed magnetic field generating unit, which includes a plurality of electromagnets arranged in an array, and is used to generate a local non-uniform gradient magnetic field, so as to realize addressable operation and group cooperative driving of the ferrofluid droplet robot; The environmental perception unit detects the real-time position information of the ferrofluid droplet robot and obstacles through the visual system, and performs multi-target position detection, tracking and positioning of the ferrofluid droplet robot group and environmental information perception; The motion planning unit detects and handles path conflicts based on the position and obstacle information of each ferrofluid droplet robot through a path planning algorithm, and plans a collision-free motion path for the ferrofluid droplet robot formation; A closed-loop controller adjusts the motion state of the ferrofluid droplet robot according to feedback information from the environment sensing unit, so that it moves along the path planned by the motion planning unit to adapt to different motion requirements, and dynamically controls the shape change of the droplet robot to adapt to changes in the surrounding environment to achieve closed-loop feedback control; A driving control unit connected to the closed-loop controller drives the electromagnets in the distributed magnetic field generating unit according to the control signal thereof to precisely control the coordinated movement of the ferrofluid droplet robot group; The motion planning unit uses incremental conflict search path planning to perform path planning in a dynamically changing working environment, specifically including: A conflict information retention unit, used to retain previous search information, including path conflict information, when the environment or task changes; The initial path planning subunit is used to calculate the initial paths for all robots in the initial situation and detect conflicts, dividing the problem into high-level conflict detection and low-level single robot path search; Dynamic change detection unit, used to detect changes in the environment or task, including new obstacles, changes in the number of robots, and changes in the target point, and identify the affected robots; Incremental conflict handling unit, used to make local adjustments to the affected robot paths, minimize the conflict area, reduce the cost of path planning and calculation, and ensure that the changed planning results are still conflict-free; Among them, the high-level conflict detection creates constraints and divides the conflict into multiple branches when a multi-robot path conflict is detected, while the low-level path search uses a classic path planning algorithm to solve the path for each robot while satisfying the constraints passed from the high-level until the paths of all robots are free of conflict.

2. The autonomous navigation control system of the reconfigurable ferrofluid droplet robot according to claim 1, characterized in that: The drive control unit includes a communicating computer and an embedded system. The computer serves as the central processing unit of the system to process control signals and generate control instructions. The embedded system receives instructions from the computer to control the current drive board to output current to modulate the distributed local magnetic field to achieve programmable movement and deformation of the droplet robot.

3. The autonomous navigation control system of the reconfigurable ferrofluid droplet robot according to claim 1, characterized in that: The distributed magnetic field generating unit includes a plurality of electromagnets arranged in an array, each electromagnet is individually connected to a power source, and when currents of different magnitudes, directions and frequencies are passed through, a specific time-varying magnetic field can be generated as required. The magnitude and range of the magnetic field are related to the applied current. The magnetic field generated by a single electromagnet is expressed as: ； in µ 0 is the vacuum permeability, N is the total number of turns of the coil, I is the current passing through the coil, M is the magnetization intensity of the iron core, z is the distance from the electromagnet surface along the z-axis in space, L is the length of the coil winding, R is the radius of the coil, and r are the length and radius of the core respectively; When the ferrofluid droplet robot is in the non-uniform gradient field generated by the electromagnet, the ferrofluid will be magnetized and thus affected by the external magnetic field. It is expressed as: ； in V represents the volume of the magnetic fluid droplet, m represents the magnetization intensity of the magnetic fluid droplet, and B represents the magnetic flux density of the external magnetic field.

4. The autonomous navigation control system of the reconfigurable ferrofluid droplet robot according to any one of claims 1 to 3, characterized in that: The driving modes of the distributed magnetic field generating unit include a single-coil driving mode and a double-coil driving mode; In the single-coil driving mode, by activating the coil adjacent to the ferrofluid droplet robot, a local magnetic field is generated to magnetize the paramagnetic droplet, thereby attracting the ferrofluid droplet robot to move to the center of the coil; In the dual-coil driving mode, by applying opposite currents to adjacent coils, the ferrofluid droplet robot is first stretched into a strip under the action of the magnetic field, and then the ferrofluid droplet robot is prompted to maintain streamlined movement by changing the activated coil. At the same time, by applying the same current to adjacent coils, the ferrofluid droplet robot is induced to split, so that it can pass through a narrow confined space smaller than itself.

5. The autonomous navigation control system of the reconfigurable ferrofluid droplet robot according to any one of claims 1 to 3, characterized in that: The environmental perception unit detects static and dynamic objects in the workspace, including obstacle positions, unstructured scene features, and motion parameters of ferrofluid droplet robots, and performs scene recognition, map construction, and multi-target tracking and positioning, specifically including: acquiring the first frame image of the camera, using a machine vision algorithm based on a support vector machine to identify and classify obstacles, feasible domains, and ferrofluid droplet robots, extracting potential path points in the workspace, boundary information of obstacles, and the number of droplet robots and the corresponding coordinates of each robot; mapping the real map into the simulation environment to construct a map model; tracking the robot through a multi-target tracking algorithm and updating its position coordinates in real time, and calibrating the position of the electromagnet to activate the corresponding coil to independently drive the droplet robot.

6. The autonomous navigation control system of the reconfigurable ferrofluid droplet robot according to any one of claims 1 to 3, characterized in that: According to the perception of environmental information, the ferrofluid droplet robot is controlled to adapt to the unstructured environment by deforming or splitting into smaller sub-droplets during movement; the movement process is divided into three states: an initial state, in which the ferrofluid droplet robot moves controllably under the drive of a single coil; a stretching state, in which the ferrofluid droplet robot deforms into a long strip and moves under the drive of multiple coils; and a splitting state, in which the ferrofluid droplet robot splits into multiple sub-droplets and moves.

7. The autonomous navigation control system of the reconfigurable ferrofluid droplet robot according to claim 6, characterized in that: The environment perception unit detects the shape of the robot during the robot's form switching process, including: The image preprocessing subunit is used to convert the color image acquired by the camera into a grayscale image, perform a Gaussian blur operation to smooth the image, and then use the Canny edge detection algorithm to obtain the edge information of the robot in the image; A feature extraction subunit is used to match the target shape with a known template using a template matching-based method, find similarities, and classify using the geometric features of the shape; The shape recognition subunit is used to recognize the shape of the ferrofluid droplet robot according to the extracted geometric features.

8. The autonomous navigation control system of the reconfigurable ferrofluid droplet robot according to any one of claims 1 to 3, characterized in that: The closed-loop controller uses a finite state machine to construct a mutual conversion strategy between various modes of the reconfigurable droplet robot, wherein a corresponding state table and action query table are established, and the state is switched according to the trigger event; wherein, when there is no obstruction on the front path, the ferrofluid droplet robot is controlled to maintain the initial state movement; when there is a confined space on the front path, the ferrofluid droplet robot is controlled to deform into a long strip movement or split into sub-droplet movement according to the spatial characteristics.

9. The autonomous navigation control system of the reconfigurable ferrofluid droplet robot according to any one of claims 1 to 3, characterized in that: The motion planning unit further comprises: The static obstacle avoidance planning subunit is used to plan a collision-free path for the robot group when only static obstacles exist, so that the robots meet the physical constraints and avoid obstacles to reach the target point; The conflict tree construction subunit is used to construct a conflict tree at a high level when a robot path conflict is detected, and generate two child nodes for each conflict in the conflict tree, each of which constrains the path of a conflicting robot to force avoidance of the conflict; The path update subunit is used to update the path using the D* path planning algorithm at the low level and continue to detect new conflicts after the update until all conflicts are resolved; The environment change response subunit is used to receive the change information transmitted by the environment perception unit when new obstacles appear in the workspace, and make local path adjustments according to the new environment map to achieve obstacle avoidance; The dynamic path adjustment subunit is used to re-plan the feasible path according to the updated environment map when multiple new obstacles are detected in the environment, so that the robot can successfully reach the target location; The static obstacle avoidance planning subunit ensures that the distance constraint between the robots is set to the length of two coils to avoid mutual interference of the magnetic fields of adjacent coils, and ensures that there is no conflict between the robots in the same time step when there is a path intersection.

Images