A control sensing device and state detection method for a soft robot

Abstract

This application discloses a control and sensing device for a soft robot and its state detection method, including a side module driving component and a corner module driving component; an air source for providing positive and negative pressure air; a solenoid valve group connected to the air source, the side module driving component, and the corner module driving component, and used for switching the air source on and off with the side module driving component and / or with the corner module driving component; a first air pressure sensor connected to the side module driving component for sensing the air pressure inside the side module driving component; a second air pressure sensor connected to the corner module driving component for sensing the air pressure inside the corner module driving component; and a control module electrically connected to the first air pressure sensor, the second air pressure sensor, and the solenoid valve group. This addresses the problems of poor flexibility and poor environmental adaptability in existing robots due to rigid control methods.

Description

Technical Field

[0001] This application relates to the field of soft robot technology, and more particularly to a control and sensing device for a soft robot and its state detection method. Background Technology

[0002] With the rise of robotics technology, its applications in fields such as intelligent manufacturing and medicine are increasing, leading to a greater diversity of robot characteristics and functions. The demand for robots to perform various tasks in different and uncertain environments is growing, thus placing higher requirements on the structural adaptability of robots. Driven by these needs, modular reconfigurable robots are finding wider application. Modular reconfigurable robots are created by combining individual robot modules in different ways, allowing them to be reconstructed into different robot forms. This endows the robots with unique and adaptive capabilities, enabling them to complete different tasks in various environments.

[0003] However, in robot control, rigid robot control methods are usually adopted, such as ball screw linear module drive, gear drive, etc., to drive rigid robots. Due to the rigid structure and rigid drive method of rigid robots, they have limited flexibility and poor adaptability to the environment in some tortuous working environments. This limits the robot's ability to complete specific tasks or in complex working environments.

[0004] Therefore, existing technologies still need to be improved and developed. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this application is to provide a control and sensing device for soft robots and a state detection method thereof, so as to solve the problems of poor flexibility and poor adaptability to the environment caused by rigid control methods in existing robots.

[0006] The technical solution of this application is as follows:

[0007] On one hand, this application proposes a control sensing device for a soft robot, wherein the control sensing device includes:

[0008] The edge module driver component and the corner module driver component are used. The edge module driver component can be linearly scaled, while the corner module driver component can be spherically expanded and contracted.

[0009] Gas source, used to provide positive pressure gas and negative pressure gas;

[0010] The solenoid valve assembly connects to the air source, the side module drive assembly, and the corner module drive assembly, and is used to switch the air source on and off with the side module drive assembly and / or with the corner module drive assembly.

[0011] The first air pressure sensor is connected to the side module drive assembly and is used to sense the air pressure inside the side module drive assembly.

[0012] The second air pressure sensor is connected to the corner module drive assembly and is used to sense the air pressure inside the corner module drive assembly.

[0013] The control module is electrically connected to the first pressure sensor, the second pressure sensor, and the solenoid valve group.

[0014] Optionally, the gas source includes: a positive pressure gas source for providing positive pressure gas; and a negative pressure gas source for providing negative pressure gas.

[0015] The solenoid valve assembly includes: a side module valve component and an angle module valve component. The side module valve component is used to connect the positive pressure air source, the negative pressure air source, and the side module drive assembly to control the extension and retraction of the side module drive assembly. The angle module valve component is used to connect the positive pressure air source, the negative pressure air source, and the angle module drive assembly to control the expansion and contraction of the angle module drive assembly.

[0016] The side module valve component includes: a first side module solenoid valve and a second side module solenoid valve; the input end of the first side module solenoid valve is connected to a positive pressure air source and a negative pressure air source respectively, the output end of the first side module solenoid valve is connected to the second side module solenoid valve, and the output end of the second side module solenoid valve is connected to the side module drive component.

[0017] The angle module valve component includes: a first angle module solenoid valve and a second angle module solenoid valve; the input end of the first angle module solenoid valve is connected to a positive pressure air source and a negative pressure air source respectively, the output end of the first angle module solenoid valve is connected to the second angle module solenoid valve, and the output end of the second angle module solenoid valve is connected to the angle module drive assembly.

[0018] Optionally, the control module includes: a voltage conversion module, which is electrically connected to the first pressure sensor and the second pressure sensor respectively;

[0019] The data acquisition module is electrically connected to the voltage conversion module.

[0020] The computing and control module is electrically connected to the data acquisition module and the air pressure output device.

[0021] Optionally, multiple edge module driver components and multiple corner module driver components can be set.

[0022] The side module drive component is connected to the side module valve component and the first air pressure sensor in a one-to-one correspondence;

[0023] The angle module drive assembly is connected one-to-one with the angle module valve component and the second air pressure sensor;

[0024] Multiple edge module driving components are set along different straight lines to form a cube frame, and multiple corner module driving components are located at the vertices of the cube frame.

[0025] Optionally, the side module driving component includes a first flexible driving part that extends and retracts linearly along a preset direction, and a first connecting part connected to the first flexible driving part;

[0026] The first flexible drive unit is connected to the side module valve component, and the first air pressure sensor is connected to the first flexible drive unit;

[0027] The corner module drive assembly includes: a second connecting part, and a second flexible drive part that can spherically expand or contract connected to the second connecting part;

[0028] The second flexible drive unit is connected to the angle module valve component, and the second air pressure sensor is connected to the second flexible drive unit;

[0029] The first connecting part and the second connecting part are detachably connected;

[0030] The side module driver component and the corner module driver component are connected by magnetic attraction between the first connecting part and the second connecting part.

[0031] On the other hand, this application also proposes a state detection method for a control sensing device of a soft robot, used in the control sensing device as described above, comprising the following steps:

[0032] Pre-inflate several side module drive components and corner module drive components respectively, and control module collects the initial air pressure of each side module drive component and each corner module drive component;

[0033] The pressure data of each side module drive component and each corner module drive component are acquired in real time by the control module. The pressure data of the side module drive component is collected by the first air pressure sensor, and the pressure data of the corner module drive component is collected by the second air pressure sensor. The pressure data includes the air pressure change.

[0034] The pressure data of the edge module driver component and the pressure data of the corner module driver component are processed by a perception algorithm to obtain the recognition results of the edge module driver component and the corner module driver component, respectively.

[0035] The state of the cubic frame soft robot in the previous stage, the recognition results of the side module driving components and the corner module driving components, as well as the pressure data of the side module driving components and the corner module driving components are input into the trained neural network model, and the current state of the cubic frame soft robot is output.

[0036] Optionally, the step of acquiring pressure data of each side module drive component and each corner module drive component in real time through the control module includes:

[0037] The voltage signals collected by the first and second air pressure sensors are filtered and detrended by the numerical control module, and the air pressure change is determined by combining the initial air pressure.

[0038] In the step of processing the pressure data of the edge module driver component and the pressure data of the corner module driver component through a perception algorithm to obtain the recognition results of the edge module driver component and the corner module driver component respectively:

[0039] The pressure data of the edge module driving component is thresholded by the sponge perception algorithm to obtain the recognition result of the edge module driving component.

[0040] The pressure data of the corner module driver component is thresholded by using a balloon sensing algorithm to obtain the recognition result of the corner module driver component;

[0041] Based on the identification results of the side module driving component and the corner module driving component, it is determined whether the soft robot is in a state of being compressed, in a state of contact with the ground, or in a state of being impacted.

[0042] Optionally, the step of performing threshold processing on the pressure data of the edge module driving component using the sponge perception algorithm to obtain the recognition result of the edge module driving component specifically includes:

[0043] The sponge sensing algorithm is used to perform threshold processing on the pressure data of the edge module driving component;

[0044] If the pressure value of one or more side module driving components exceeds a preset change pressure value within a preset time, the identification result of the side module driving component is determined to be under pressure.

[0045] Optionally, the step of performing threshold processing on the pressure data of the corner module driving component using the balloon sensing algorithm to obtain the recognition result of the corner module driving component specifically includes:

[0046] The pressure data of the diagonal module driving component is thresholded using a balloon sensing algorithm.

[0047] If the pressure value of one or more corner module drive components exceeds a preset change pressure value within a preset time, the identification result of the corner module drive component is determined to be that it is being squeezed.

[0048] Optionally, based on the identification results of the edge module driving component and the corner module driving component, the specific steps for determining whether the soft robot is in a state of being compressed, in a state of contact with the ground, or in a state of being impacted include:

[0049] When the edge module driving component is identified as being compressed, the soft robot is determined to be in a compressed state. Based on the air pressure change of one or more edge module driving components, the size and location of the external object compressing the soft robot are determined.

[0050] When the corner module drive component is identified as being squeezed, the pressure change of one or more corner module drive components and the previous state of the soft robot are used to determine whether the three-dimensional surface of the soft robot in contact with the ground has changed.

[0051] When the edge module driving component is identified as being compressed and the corner module driving component is identified as being squeezed, it is determined that the soft robot has been impacted, and the direction of the impact is determined.

[0052] Beneficial Effects: Compared with existing technologies, the control and sensing device and its state detection method for soft robots proposed in this application involve the control and sensing device collecting air pressure data from the side and corner drive components of the soft robot, converting it into pressure signals, and processing these signals with the control module to obtain identification results for several side and corner drive components. Based on the previous state of the soft robot and the pressure data and identification results of these components, the current state of the soft robot can be detected, thereby enabling the detection of whether the soft robot has experienced flipping, compression, or collisions. The control of the side and corner drive components is achieved by controlling the on / off state of the air source and the drive components via a solenoid valve group. The soft robot formed using this control and sensing device has a simple structure, low cost, and is easy to operate. Compared with traditional rigid robots, it can complete different tasks in different environments, adapt to task execution in more variable environments, and has better performance. Furthermore, the main components of the linearly scalable side module drive assembly and the spherically expandable corner module drive assembly are made of soft materials. This results in a soft robot with higher compliance and structural variability, better environmental adaptability and degrees of freedom, enabling the soft robot to complete various tasks in complex environments. Attached Figure Description

[0053] Figure 1 This is a schematic block diagram of the circuit and air path of the control and sensing device for a soft robot according to an embodiment of this application.

[0054] Figure 2 This is a schematic diagram of the cubic frame of the control and sensing device for a soft robot according to an embodiment of this application.

[0055] Figure 3 This is a cross-sectional view of the side module driving component of the control and perception device for a soft robot according to an embodiment of this application;

[0056] Figure 4 This is a schematic diagram of the corner module drive assembly of the control and sensing device for a soft robot according to an embodiment of this application;

[0057] Figure 5 This is a schematic diagram of the force state of the cubic frame of the control and sensing device for a soft robot according to an embodiment of this application;

[0058] Figure 6 This is a schematic diagram illustrating the changing state of the control and sensing device of the soft robot in an embodiment of this application when the driving component of the sensing edge module is subjected to force.

[0059] Figure 7 This is a schematic diagram illustrating the changing state of the control and sensing device of the soft robot in an embodiment of this application when the driving component of the sensing angle module is subjected to force.

[0060] Figure 8 This is a schematic diagram illustrating the change in state of the control and sensing device of a soft robot in an embodiment of this application when sensing an impact.

[0061] Figure 9 This is a main flowchart of a state detection method for a control sensing device of a soft robot according to an embodiment of this application;

[0062] Figure 10 This is a detailed flowchart illustrating a state detection method for a control sensing device of a soft robot according to an embodiment of this application.

[0063] Figure 11 This is a data processing flowchart of a state detection method for a control and sensing device of a soft robot, according to an embodiment of this application.

[0064] The following labels are used in the diagram: 100, Air source; 110, Positive pressure air source; 120, Negative pressure air source; 200, Solenoid valve assembly; 210, Side module valve component; 211, First side module solenoid valve; 212, Second side module solenoid valve; 220, Angle module valve component; 221, First angle module solenoid valve; 222, Second angle module solenoid valve; 310, Side module drive assembly; 311, First flexible drive unit; 312, Flexible housing; 313, Flexible support body; 314, First air pressure. 315. Conveying pipe; 316. First connecting part; 320. First magnetic suction component; 321. Corner module driving assembly; 322. Second flexible driving part; 323. Driving airbag; 324. Second air pressure conveying pipe; 325. Second connecting part; 326. Base frame; 410. Second magnetic suction component; 420. First air pressure sensor; 500. Second air pressure sensor; 510. Control module; 520. Voltage conversion module; 530. Data acquisition module; 530. Calculation and control module. Detailed Implementation

[0065] This application provides a control and sensing device for soft robots and a state detection method thereof. To make the objectives, technical solutions, and effects of this application clearer and more explicit, the following describes the application in optional detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0066] Example 1

[0067] like Figure 1As shown, this embodiment proposes a control and sensing device for a soft robot, including an edge module driving component 310 and a corner module driving component 320. The edge module driving component 310 is linearly expandable and contractable, while the corner module driving component 320 is spherically expandable and contractable. A polyhedral frame, such as a cubic frame, is constructed using the edge module driving component 310 and the corner module driving component 320. The edge module driving component 310 is located at the edge line of the polyhedral frame, and multiple corner module driving components 320 are located at the vertices of the polyhedral frame. The control and sensing device also includes: an air source 100, a solenoid valve group 200, a first air pressure sensor 410, a second air pressure sensor 420, and a control module 500. The air source 100 is used to provide positive and negative pressure air, providing driving force for the linear expansion and contraction of the edge module driving component 310 and the spherical expansion and contraction of the corner module driving component 320. The solenoid valve assembly 200 connects to the air source 100, the side module drive assembly 310, and the corner module drive assembly 320. The solenoid valve assembly 200 controls the connection and disconnection between the air source 100 and the side module drive assembly 310, and / or the connection and disconnection between the air source 100 and the corner module drive assembly 320. Thus, under the control of the solenoid valve assembly 200, the side module drive assembly 310 extends when positive pressure is applied and shortens when negative pressure is applied; similarly, the corner module drive assembly 320 expands when positive pressure is applied and contracts when negative pressure is applied. A first pressure sensor 410 is connected to the side module drive assembly 310 and is used to sense the air pressure within it. A second pressure sensor 420 is connected to the corner module drive assembly 320 and is used to sense the air pressure within it. The control module 500 is electrically connected to the first pressure sensor 410, the second pressure sensor 420, and the solenoid valve assembly 200.

[0068] In this embodiment, the solenoid valve group 200 controls the on / off connection between the side air source 100 and the module drive component, and / or the air source 100 and the corner module drive component 320. This allows control of the side module drive component 310 and the corner module drive component 320 via the input of positive and negative pressure air. Furthermore, stopping the air supply keeps the positions of the side module drive component 310 and the corner module drive component 320 stationary, thus stopping the soft robot. Using this control method, the air source 100 and the solenoid valve group 200 are located outside the soft robot, and the gas is typically connected via flexible hoses, thus achieving flexible control of the soft robot's movements. Furthermore, the first air pressure sensor 410 senses the air pressure of the side module drive component 310, and the second air pressure sensor 420 senses the air pressure of the corner module drive component 320. After receiving the air pressure data, the control module 500 can process and analyze it, and control the solenoid valve group 200 through the control module 500 to form a control feedback structure, which facilitates timely adjustment of the control of the side module drive component 310 and the corner module drive component 320. By controlling the sensing devices to collect the air pressure data of the side module drive component 310 and the corner module drive component 320 of the soft robot, converting them into pressure signals, and processing them by the control module 500 to obtain the identification results of several side module drive component 310 and corner module drive component 320, based on the previous stage state of the soft robot and the pressure data and identification results of several side module drive component 310 and corner module drive component 320, the current state of the soft robot can be detected, thereby realizing the detection of whether the soft robot has flipped, been compressed, or collided, thus realizing the detection of the soft robot. The soft robot formed using this control and sensing device has a simple structure, low cost, and easy operation. Compared with traditional rigid robots, it can complete different tasks in different environments, adapt to task execution in more variable environments, and has better performance. Moreover, the main parts of the linearly stretchable side module drive component 310 and the spherically expandable corner module drive component 320 are made of soft materials, which makes the soft robot structure more compliant and structurally variable, with better environmental adaptability and degrees of freedom, enabling the soft robot to complete various tasks in complex environments.

[0069] like Figure 1As shown, the air source 100 in this embodiment specifically includes a positive pressure air source 110 and a negative pressure air source 120. The positive pressure air source 110 is used to provide positive pressure air (high pressure), and the negative pressure air source 120 is used to provide negative pressure air (high pressure). The solenoid valve assembly 200 specifically includes a side module valve component 210 and a corner module valve component 220. The side module valve component 210 is used to connect the positive pressure air source 110, the negative pressure air source 120, and the side module drive assembly 310 to control the extension and retraction of the side module drive assembly 310. The corner module valve component 220 is used to connect the positive pressure air source 110, the negative pressure air source 120, and the corner module drive assembly 320 to control the expansion and contraction of the corner module drive assembly 320. When driving the soft robot through the air source 100 and the solenoid valve assembly 200, the air source 100 and the solenoid valve assembly 200 can be placed outside the soft robot and connected by a flexible hose. The length of the hose can be set according to the needs of the working environment. By placing the air source 100 and the solenoid valve assembly 200 outside the working area, the soft robot has greater freedom of movement, making it more suitable for various working environments.

[0070] like Figure 1 As shown, the side module valve component 210 in this embodiment includes: a first side module solenoid valve 211 and a second side module solenoid valve 212; the input end of the first side module solenoid valve 211 is connected to the positive pressure air source 110 and the negative pressure air source 120 respectively, the output end of the first side module solenoid valve 211 is connected to the second side module solenoid valve 212, and the output end of the second side module solenoid valve 212 is connected to the side module drive assembly 310. The angle module valve component 220 includes: a first angle module solenoid valve 221 and a second angle module solenoid valve 222; the input end of the first angle module solenoid valve 221 is connected to the positive pressure air source 110 and the negative pressure air source 120 respectively, the output end of the first angle module solenoid valve 221 is connected to the second angle module solenoid valve 222, and the output end of the second angle module solenoid valve 222 is connected to the angle module drive assembly 320. The first-side module solenoid valve 211 and the first-angle module solenoid valve 221 function identically to switch between positive and negative pressure input air, thereby controlling the extension or shortening. The second-side module solenoid valve 212 and the second-angle module solenoid valve 222 function identically to open or close the air path, thereby starting or stopping the air supply. Using two solenoid valves for control simplifies the control method and optimizes the control structure.

[0071] Alternatively, it is conceivable that the side module valve component 210 and the angle module valve component 220 could also employ other types of solenoid valves. For example, multi-way multi-control solenoid valves, where a single solenoid valve can achieve the functions of supplying positive pressure air, supplying negative pressure air, and shutting off air supply through different control positions.

[0072] Furthermore, such as Figure 1As shown, the control module 500 in this embodiment specifically includes: a voltage conversion module 510, a data acquisition module 520, and a calculation and control module 530. The voltage conversion module 510 is electrically connected to the first air pressure sensor 410 and the second air pressure sensor 420, respectively, and is used to amplify the voltage values ​​sensed by the first air pressure sensor 410 and the second air pressure sensor 420. The data acquisition module 520 is electrically connected to the voltage conversion module 510 and is used to acquire and process all the air pressure values ​​sensed by the first air pressure sensor 410 and the second air pressure sensor 420. The calculation and control module 530 is electrically connected to both the data acquisition module 520 and the air pressure output device. The calculation and control module 530 is typically a processor. Through data processing and analysis, the processor calculates the current pressure data based on corresponding linear relationships, thereby outputting control signals (e.g., control commands for the solenoid valve group 200) or drawing relevant conclusions (e.g., judgment of the robot's state).

[0073] Furthermore, such as Figure 1 , Figure 2 , Figure 3 As shown, multiple edge module driving components 310 and multiple corner module driving components 320 are provided. The multiple edge module driving components 310 are arranged along different straight lines to form a polyhedral frame, and the multiple corner module driving components 320 are located at the vertices of the polyhedral frame. In this embodiment, the polyhedral frame is a cube frame. The multiple edge module driving components 310 in the cube frame are arranged along the up-down, left-right, and front-back directions, respectively; the multiple corner module driving components 320 are located at the vertices of the cube frame. By using a cube frame, the lengths of the edge module driving components 310 on each side of the cube frame are equal, and the sizes of the corner module driving components 320 at each corner are the same. This makes it easier to form reconfigurable modules, which is beneficial for the precise control of the flexible robot. One edge module driving component 310 in the cube frame is connected to an edge module valve component 210 and a first pressure sensor 410; one corner module driving component 320 in the cube frame is connected to a corner module valve component 220 and a second pressure sensor 420. This allows for a one-to-one connection, enabling each side module driver component 310 and corner module driver component 320 to be individually sensed and controlled.

[0074] Furthermore, such as Figure 2 , Figure 3As shown, the side module driving assembly 310 includes a first flexible driving part 311 that extends and retracts linearly along a preset direction, and a first connecting part 315 connected to the first flexible driving part 311. The first flexible driving part 311 is connected to the side module valve component 210, and a first air pressure sensor 410 is connected to the first flexible driving part 311. The corner module driving assembly 320 includes a second connecting part 324, and a second flexible driving part 321 that is spherically expandable or contractible and connected to the second connecting part 324. The second flexible driving part 321 is connected to the corner module valve component 220, and a second air pressure sensor 420 is connected to the second flexible driving part 321. The side module driving assembly 310 and the corner module driving assembly 320 are connected by magnetic attraction between the first connecting part 315 and the second connecting part 324. The edge module drive component 310 and the corner module drive components 320 at both ends are detachably connected via a first connecting part 315 and a second connecting part 324, allowing for disassembly and reassembly. Furthermore, the polyhedral frame can be connected to other polyhedral frames to form different soft robot structures, thus achieving reconfiguration. Moreover, the soft robot formed by using a linearly extensible first flexible drive part 311 and a spherically expanding or contracting second flexible drive part 321 as its main structure is simple in structure, low in cost, and easy to operate. Compared with traditional rigid robots, it can complete different tasks in different environments, adapting to task execution in more varied environments and exhibiting better performance. Furthermore, the main parts of the first flexible drive part 311 and the second flexible drive part 321 are made of soft materials, resulting in a robot with higher compliance and structural variability, exhibiting better environmental adaptability and degrees of freedom.

[0075] The first flexible drive unit 311 specifically includes: a flexible shell 312, a flexible support 313, and a first pneumatic conveying pipe 314. The flexible shell 312 has an air cavity extending in a preset direction, which is a straight line. The flexible support 313 is disposed within the air cavity and extends in the preset direction. In this embodiment, the flexible support 313 can be made of sponge, which has a certain supporting strength and can be compressed and stretched, providing stable support for the flexible shell 312. The flexible support 313 can also be made of other materials, such as springs with a certain degree of plasticity or other flexible materials that can provide support. The flexible shell 312 is made of a sealable plastic material to encapsulate the flexible support 313, and may have pores. The first pneumatic conveying pipe 314 connects to the air cavity through the pores. The rest of the air cavity within the flexible shell 312 is completely sealed. The first pneumatic conveying pipe 314 connects to the side module valve component 210 to provide positive and negative air pressure, enabling the flexible shell 312 to linearly expand and contract. The first air pressure sensor 410 is connected to the air cavity of the flexible housing 312 to sense the air pressure inside the flexible housing 312.

[0076] In this embodiment, the first connecting portion 315 specifically includes at least two first magnetic suction members 316, which are respectively disposed within the flexible housing 312 and located at both ends of the flexible support 313. By disposing of the first magnetic suction members 316 within the flexible housing 312, the flexible housing 312 can limit the first magnetic suction members 316, ensuring the stability of the installation.

[0077] like Figure 1 , Figure 2 , Figure 4 As shown, the second flexible drive unit 321 in this embodiment specifically includes a drive airbag 322 and a second air pressure delivery pipe 323. The drive airbag 322 is connected to the second connecting part 324, which allows it to dock with the first connecting part 315 provided on the first flexible drive unit 311. The drive airbag 322 can be a round latex balloon. One end of the second air pressure delivery pipe 323 is connected to the drive airbag 322, and the other end is connected to the angle module valve component 220. It is connected to the air source 100 to provide positive and negative air pressure to inflate or contract the drive airbag 322.

[0078] like Figure 2 , Figure 4 As shown, the second connecting part 324 in this embodiment specifically includes a base frame 325 and a second magnetic member 326. The driving airbag 322 is disposed within the base frame 325, and an opening is provided on the base frame 325 so that the driving airbag 322 can protrude from the opening. This prevents excessive interference from the base frame 325 during the expansion of the driving airbag 322. The second magnetic member 326 is disposed on the base frame 325 and is used to generate an attractive force with the first magnetic member 316. When the first flexible driving part 311 and the second flexible driving part 321 are connected, the docking can be achieved directly through the attractive force between the first magnetic member 316 and the second magnetic member 326. This connection method is simple and facilitates disassembly and reconfiguration. It offers greater flexibility, allowing the soft robot to be adjusted promptly according to the usage scenario.

[0079] like Figure 1 , Figure 2 , Figure 5 As shown, the soft robot in this embodiment can be formed by magnetically attaching multiple cube frames. When the cube frames are subjected to external force, the edge module driving component 310 deforms under pressure, thereby sensing the air pressure change of the edge module driving component 310 through the first air pressure sensor 410. The corner module driving component 320 deforms under pressure, and the second air pressure sensor 420 senses the air pressure change of the corner module driving component 320. By processing and analyzing these air pressure changes, the state of the soft robot can be determined.

[0080] Example 2

[0081] like Figure 9 As shown, this embodiment proposes a state detection method for a control sensing device of a soft robot, used in the control sensing device described in Embodiment 1. The state detection method includes the following steps:

[0082] Step S100: Pre-inflate several side module drive components and corner module drive components respectively, and control module collects the initial air pressure of each side module drive component and each corner module drive component.

[0083] In the specific process, appropriate pre-pressure is applied to several side module drive components and corner module drive components. The pressure at this time is collected by the control module and used as the initial pressure of each side module drive component and corner module drive component.

[0084] Step S200: The pressure data of each side module drive component and each corner module drive component are acquired in real time through the control module. The pressure data of the side module drive component is collected by the first air pressure sensor, and the pressure data of the corner module drive component is collected by the second air pressure sensor. The pressure data includes the air pressure change.

[0085] Several pressure sensors convert the pressure signals from the side module drive components into electrical signals, and the pressure signals from the corner module drive components into electrical signals. These signals are then fed back to the voltage conversion module in real time. The voltage conversion module amplifies and processes the electrical signals. The data acquisition module acquires and processes the voltage signals to obtain the pressure change. The calculation and control module calculates the pressure data at this time using the corresponding linear relationship.

[0086] The specific steps of step S200 in this embodiment include: filtering and detrending the voltage signals collected by the first and second pressure sensors respectively through the digital control module, and determining the pressure change by combining the initial pressure.

[0087] Step S300: The pressure data of the edge module driving component and the pressure data of the corner module driving component are processed by the perception algorithm to obtain the recognition results of the edge module driving component and the corner module driving component, respectively.

[0088] After determining the pressure change by combining the initial air pressure, the sensing algorithm can detect whether the side module drive component is being compressed, and whether the corner module drive component is in contact with the ground.

[0089] like Figure 9 , Figure 10 As shown, step S300 specifically includes:

[0090] Step S310: Threshold processing is performed on the pressure data of the edge module driving component using the sponge perception algorithm to obtain the identification result of the edge module driving component. For example... Figure 6 The diagram shows (Figure E: no force applied, Figure F: under pressure applied) illustrating the process of a soft robot being compressed by an external object. The specific identification process is as follows: A sponge sensing algorithm is used to threshold the pressure data of the side module drive components. If the pressure values ​​of one or more side module drive components exceed a preset pressure change value within a preset time period (e.g., a large pressure change is sensed in a very short time, with a sudden change value of approximately 1 kPa), this method can detect sudden changes in the pressure values ​​of multiple side module drive components. Based on these sudden changes, the identification result of the side module drive component is determined to be under compression. The degree of compression identification can then be used to determine the state of external pressure exerted on the soft robot.

[0091] Step S320: Threshold processing is performed on the pressure data of the corner module driver component using a balloon sensing algorithm to obtain the identification result of the corner module driver component. For example... Figure 7 As shown in the diagram (A: initial air pressure state; B: state of the contracted top corner module drive component; C: state of detecting sudden air pressure changes; D: state of the expanded corner module drive component, determining the bottom surface state), the process of the soft robot's corner module drive component being compressed is as follows: A balloon sensing algorithm is used to threshold the pressure data of the corner module drive component. If the pressure value of one or more corner module drive components exceeds a preset pressure change value within a preset time, i.e., a large pressure change is sensed in a very short time (e.g., a sudden change of about 1 kPa), this method can detect sudden changes in the pressure values ​​of multiple corner module drive components. The identification result of the corner module drive component is determined to be that it is being compressed. Based on the degree of compression, the state of the soft robot being compressed by its own overturning can be determined.

[0092] Step S330: Based on the recognition results of the side module driving component and the corner module driving component, determine whether the soft robot is in a state of being compressed, in a state of contact with the ground, or in a state of being impacted.

[0093] The specific judgment process is as follows: Figure 6 As shown, when the recognition result of the side module driving component being compressed is determined, the software robot is impacted and the side module driving component is deformed by force, which leads to changes in the air pressure inside the side module. It is determined that the soft robot is being squeezed by an external object and is in a compressed state. Based on the amount of air pressure change of one or more side module driving components, the size and location of the external object squeezing the soft robot are determined.

[0094] like Figure 7As shown, when the corner module driving component is identified as being squeezed, the software robot flips its direction, causing the corner module driving component to be deformed under pressure. This results in a change in the air pressure inside the corner module. Based on the amount of air pressure change in one or more corner module driving components and the previous state of the soft robot, it is determined whether the three-dimensional surface of the soft robot in contact with the ground has changed.

[0095] like Figure 8 As shown in Figure G (where the robot is not subjected to impact and Figure H is subjected to impact), when the side module driving component is identified as being compressed and the corner module driving component is identified as being squeezed, the robot is determined to have been impacted, and the direction of the impact is determined.

[0096] Step S400: Input the state of the soft robot in the previous stage, the recognition results of the side module driving component and the corner module driving component, as well as the pressure data of the side module driving component and the pressure data of the corner module driving component into the trained neural network model, and output the current stage state of the soft robot.

[0097] like Figure 11 As shown, the neural network model can be pre-trained and use multiple sets of pre-tested experimental data as samples to enable the neural network model to identify different force conditions and output the current state of the soft robot, thus achieving accurate identification of the current state of the soft robot.

[0098] The aforementioned state detection method enables the detection of whether a cubic frame soft robot has experienced flipping, compression, or collisions. The detection is accurate and the control is convenient.

[0099] In summary, this application proposes a control and sensing device and its state detection method for a soft robot. The control and sensing device collects air pressure data from the side module drive components and corner module drive components of the soft robot, converts them into pressure signals, and controls the module to process them to obtain identification results of several side module drive components and corner module drive components. Based on the previous state of the soft robot and the pressure data and identification results of several side module drive components and corner module drive components, the current state of the soft robot can be detected, thereby realizing the detection of whether the soft robot has flipped, been compressed, or collided, thus achieving the detection of the soft robot.

[0100] It should be understood that the application of this application is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A control and sensing device for a soft robot, characterized in that, include: The side module driving component and the corner module driving component are provided, wherein the side module driving component is linearly expandable and the corner module driving component is spherically expandable and contractable. A gas source, which provides positive pressure gas and negative pressure gas; A solenoid valve assembly, which is connected to the air source, the side module drive assembly, and the corner module drive assembly, and is used to switch the air source on and off with the side module drive assembly and / or the air source on and off with the corner module drive assembly. A first air pressure sensor is connected to the side module drive assembly and is used to sense the air pressure inside the side module drive assembly. A second air pressure sensor is connected to the corner module drive assembly and is used to sense the air pressure inside the corner module drive assembly; A control module, wherein the control module is electrically connected to the first pressure sensor, the second pressure sensor and the solenoid valve group respectively; The gas source includes: a positive pressure gas source, which is used to provide positive pressure gas; and a negative pressure gas source, which is used to provide negative pressure gas. The solenoid valve assembly includes: a side module valve component and an angle module valve component. The side module valve component is used to connect the positive pressure air source, the negative pressure air source, and the side module drive assembly to control the extension and retraction of the side module drive assembly. The angle module valve component is used to connect the positive pressure air source, the negative pressure air source, and the angle module drive assembly to control the expansion and contraction of the angle module drive assembly. The side module valve component includes: a first side module solenoid valve and a second side module solenoid valve; the input end of the first side module solenoid valve is connected to the positive pressure air source and the negative pressure air source respectively, the output end of the first side module solenoid valve is connected to the second side module solenoid valve, and the output end of the second side module solenoid valve is connected to the side module drive assembly. The angle module valve component includes: a first angle module solenoid valve and a second angle module solenoid valve; the input end of the first angle module solenoid valve is connected to the positive pressure air source and the negative pressure air source respectively, the output end of the first angle module solenoid valve is connected to the second angle module solenoid valve, and the output end of the second angle module solenoid valve is connected to the angle module drive assembly; Multiple edge module driving components are provided, and multiple corner module driving components are provided. The multiple edge module driving components are respectively arranged along different straight line directions to form a cube frame, and the multiple corner module driving components are respectively located at the vertices of the cube frame. The side module drive component is connected to the side module valve component and the first air pressure sensor in a one-to-one correspondence; The angle module drive component is connected to the angle module valve component and the second air pressure sensor in a one-to-one correspondence; The control module processes the identification results of several side module driving components and corner module driving components. Based on the previous stage state of the soft robot and the pressure data and identification results of several side module driving components and corner module driving components, the current state of the soft robot can be detected, thereby realizing the detection of whether the soft robot has flipped, been compressed, or collided, thus realizing the detection of the soft robot.

2. The control and sensing device for a soft robot according to claim 1, characterized in that, The control module includes a voltage conversion module, which is electrically connected to the first air pressure sensor and the second air pressure sensor respectively. The data acquisition module is electrically connected to the voltage conversion module; The calculation and control module is electrically connected to both the data acquisition module and the air pressure output device.

3. The control and sensing device for a soft robot according to claim 1, characterized in that, The edge module driving component includes a first flexible driving part that extends and retracts linearly along a preset direction, and a first connecting part connected to the first flexible driving part. The first flexible drive unit is connected to the side module valve component, and the first air pressure sensor is connected to the first flexible drive unit; The corner module driving component includes: a second connecting part, and a second flexible driving part that can spherically expand or contract connected to the second connecting part; The second flexible drive unit is connected to the angle module valve component, and the second air pressure sensor is connected to the second flexible drive unit; The side module driving component and the corner module driving component are connected by magnetic attraction between the first connecting part and the second connecting part.

4. A method for state detection of a control sensing device for a soft robot, characterized in that, For use in the control sensing device as described in any one of claims 1-3, the steps include: Pre-inflate several side module drive components and corner module drive components respectively, and control module collects the initial air pressure of each side module drive component and each corner module drive component; The pressure data of each side module drive component and each corner module drive component are acquired in real time by the control module. The pressure data of the side module drive component is collected by the first air pressure sensor, and the pressure data of the corner module drive component is collected by the second air pressure sensor. The pressure data includes the air pressure change. The pressure data of the edge module driver component and the pressure data of the corner module driver component are processed by a perception algorithm to obtain the recognition results of the edge module driver component and the corner module driver component, respectively. The state of the soft robot in the previous stage, the recognition results of the side module driving components and the corner module driving components, as well as the pressure data of the side module driving components and the corner module driving components are input into the trained neural network model, and the current state of the soft robot is output.

Images