Controlling method and system for continuum robot based on global tactile perception

Abstract

This invention provides an adaptive control method and system for continuum robots based on global tactile perception, relating to the field of continuum robots. By using a conformally attached flexible array pressure sensor, it directly measures the contact force across the entire robot body, overcoming the perception blind zone problem of existing end-effector force sensors. It also avoids the high cost and increased stiffness issues of fiber optic grating solutions, achieving low-cost, highly flexible global force perception. The heterogeneous modulus structure design effectively solves the problem of measurement inaccuracies by flexible sensors under large deformation environments. The adaptive admittance control strategy enables the robot to dynamically adjust its compliance characteristics based on real-time tactile information, significantly improving operational safety in sensitive environments (such as surgery). The sensor scheme and control framework have good universality and can be applied to continuum robot platforms with different structures.

Description

Technical Field

[0001] This invention relates to the field of robotics, and in particular to an adaptive control method and system for a continuum robot based on global tactile perception. Background Technology

[0002] Continuum robots, due to their high flexibility and continuous bending capabilities, have enormous application potential in confined, unstructured environments such as minimally invasive surgery, disaster relief, and industrial inspection. However, the continuity and flexibility of their structure also present significant challenges to the integration of physical sensors, and existing force sensing solutions have inherent limitations.

[0003] Indirect force estimation based on computer vision calculates contact force by analyzing tissue deformation through images and combining it with mechanical models. Its drawbacks are: body fluids in the surgical environment can easily cause image blurring, resulting in large calculation errors; and it requires the construction of special models for different tissues, resulting in poor universality.

[0004] Strain measurement based on fiber Bragg gratings indirectly calculates force by detecting fiber wavelength shift. Its bottlenecks are: high system cost; complex calibration of strain-to-force conversion; and fiber optic implantation increases the overall stiffness of the robot and weakens its compliance.

[0005] The solution of end-effector integrated force sensor only integrates a miniature force sensor at the end of the robot. Its shortcomings are: it cannot sense the contact force in the middle or near end of the robot body, resulting in a sensing blind spot; the introduction of rigid sensors will cause a sudden change in local stiffness, which can easily lead to stress concentration when interacting with fragile tissues.

[0006] In summary, all existing technologies suffer from a core unresolved problem: the lack of an effective means for low-cost, distributed, and direct force measurement across the entire surface of a continuum robot. Furthermore, existing control strategies are mostly based on position control or admittance control with fixed parameters, making it difficult to adaptively adjust to complex interactive states. Summary of the Invention

[0007] To achieve the above-mentioned objectives and other advantages of the present invention, the first objective of the present invention is to provide an adaptive control method for a continuum robot based on full-domain tactile perception. The continuum robot has a flexible array pressure sensor attached to its outer surface. This flexible array pressure sensor is a sensor with heterogeneous modulus partitions, where the high-modulus sensing region is attached to the rigid section of the continuum robot, and the low-modulus tensile region is attached to the deformable section of the continuum robot. This ensures the continuity and stability of the pressure measurement signal under all postures of the continuum robot. The method includes the following steps:

[0008] Acquire the two-dimensional pressure distribution data output by the flexible array pressure sensor;

[0009] Based on the kinematic model of the continuum robot, the two-dimensional pressure distribution data is mapped to the three-dimensional operating space to determine the three-dimensional position of the contact force.

[0010] Based on the determined contact force magnitude and distribution pattern, the smoothing parameters used for admittance control are dynamically adjusted.

[0011] Using the adjusted smoothing parameters, motion control commands for the continuum robot body are generated to achieve compliant interaction.

[0012] Furthermore, the step of mapping the two-dimensional pressure distribution data to the three-dimensional operating space includes:

[0013] For each sensing unit that outputs a pressure signal, determine its position in the local coordinate system of the continuum robot body;

[0014] Based on the current joint state of the continuum robot, calculate the homogeneous transformation matrix from the base coordinate system of the continuum robot to the local coordinate system where the sensing unit is located, using the following formula:

[0015]

[0016] in, Represents the sine function , Represents the cosine function , , , The length of the continuum robot;

[0017] By using spatial mapping relationships, the three-dimensional position of the sensing unit in the world coordinate system is calculated, thereby associating each pressure signal with a three-dimensional spatial contact point. The formula is as follows:

[0018]

[0019] in, Let be the homogeneous transformation matrix from the base to the end. The position of the sensing unit in the local coordinate system. This represents the position of the sensing unit in the Cartesian coordinate system.

[0020] Furthermore, the smoothing parameters include the desired virtual stiffness value and the desired virtual damping value;

[0021] The adjustment strategy for the smoothing parameters used in admittance control is as follows: the greater the contact force, the smaller the desired virtual stiffness value, and the greater the desired virtual damping value.

[0022] Furthermore, the step of dynamically adjusting the smoothing parameters used for admittance control based on the determined contact force magnitude and distribution pattern includes:

[0023] Set a pressure threshold and a safety threshold, wherein the safety threshold is greater than the pressure threshold;

[0024] When the pressure of all sensing units is less than the pressure threshold, it is determined to be a non-contact state. The desired virtual stiffness value is adjusted to the first virtual stiffness range, and the desired virtual damping value is adjusted to the first virtual damping range to ensure positioning accuracy and system response.

[0025] When the pressure of any sensing unit exceeds the pressure threshold and the pressure of all sensing units is less than the safety threshold, it is determined to be a slight contact state. The desired virtual stiffness value is adjusted to the second virtual stiffness range, and the desired virtual damping value is adjusted to the second virtual damping range to enhance the robot's compliance and interaction smoothness. The second virtual stiffness range is smaller than the first virtual stiffness range, and the second virtual damping range is larger than the first virtual damping range.

[0026] When the pressure of any sensing unit exceeds the safety threshold, a collision risk state is determined. The desired virtual stiffness value is adjusted to within the third virtual stiffness range, and simultaneously, the desired virtual damping value is adjusted to within the third virtual damping range to achieve rapid impact absorption and active avoidance. The third virtual stiffness range is smaller than the second virtual stiffness range, and the third virtual damping range is larger than the second virtual damping range. Further, after the step of dynamically adjusting the smoothing parameters used for admittance control based on the determined contact force magnitude and distribution pattern, the following is also included:

[0027] The desired virtual stiffness value and the desired virtual damping value are smoothed to obtain the smoothed virtual stiffness value and smoothed virtual damping value actually used in the current control cycle.

[0028] Furthermore, the smoothing process for the desired virtual stiffness value and the desired virtual damping value is specifically achieved through a first-order low-pass filter to achieve a smooth transition, as shown in the formula:

[0029]

[0030] in, and The virtual stiffness and virtual damping at the current moment, and For the desired virtual stiffness and desired virtual damping determined, and These are the virtual stiffness and virtual damping values ​​from the previous control cycle. It is a smoothing factor, and .

[0031] Furthermore, during the adjustment of the smoothing parameters, ensure that the system stability conditions are always met: ,in The virtual inertial mass of the admittance controller.

[0032] Furthermore, the step of generating motion control commands for the continuum robot body using the adjusted smoothing parameters to achieve compliant interaction includes:

[0033] Substituting the smoothed virtual stiffness value, the smoothed virtual damping value, and the current contact force into the admittance model, the motion correction amount of the continuum robot is calculated, and the final control command is generated accordingly to drive the continuum robot to achieve compliant interaction with the environment.

[0034] Furthermore, the admittance model is as follows:

[0035]

[0036] in, For virtual inertial mass, This is the position correction amount. For speed correction, For acceleration correction, This is the current contact force vector.

[0037] A second objective of this invention is to provide an adaptive control system for a continuum robot based on global tactile perception, which utilizes the aforementioned method and includes:

[0038] The flexible continuum robot body adopts a multi-segment continuous bending structure to continuously deform in three-dimensional space;

[0039] A flexible array pressure sensor is attached to the outer surface of the flexible continuum robot body to measure the pressure distribution on the outer surface in real time.

[0040] The signal processing and control module is electrically connected to the actuators of the flexible array pressure sensor and the flexible continuum robot body.

[0041] The flexible array pressure sensor adopts a heterogeneous modulus layered structure, which includes a high modulus sensing region and a low modulus tensile region. The high modulus sensing region is attached to the rigid section of the robot body, and the low modulus tensile region is attached to the flexible section of the robot body.

[0042] The signal processing and control module is configured to: receive pressure distribution data from the flexible array pressure sensor; map the two-dimensional pressure distribution data output by the flexible array pressure sensor to a three-dimensional operating space based on the kinematic model of the robot body to determine the three-dimensional position of the contact force; dynamically adjust the smoothing parameters used for admittance control according to the determined contact force magnitude and distribution pattern; and generate motion control commands for the continuum robot body using the adjusted smoothing parameters to achieve compliant interaction.

[0043] Furthermore, the heterogeneous modulus layered structure of the flexible array pressure sensor comprises, from top to bottom:

[0044] First flexible substrate layer;

[0045] The first stretchable conductor layer is formed of liquid metal material;

[0046] Piezoresistive sensing layer;

[0047] The second stretchable wire layer is formed of liquid metal material and is arranged to intersect with the first stretchable wire layer to define the sensing unit array.

[0048] Second flexible substrate layer;

[0049] The piezoresistive sensing layer covers the intersection of the first stretchable wire layer and the second stretchable wire layer at the position corresponding to the high modulus sensing region, but does not cover the intersection at the position corresponding to the low modulus stretchable region.

[0050] Furthermore, the first flexible substrate layer and the second flexible substrate layer are made of thermoplastic polyurethane elastomer; the liquid metal material is gallium indium alloy; and the piezoresistive sensing layer is made of Velostat material.

[0051] Furthermore, the fabrication of the flexible array pressure sensor includes the following steps:

[0052] Provide a first flexible substrate layer and a second flexible substrate layer;

[0053] On the first flexible substrate layer, a first stretchable conductive layer is patterned using a screen printing process; on the second flexible substrate layer, a second stretchable conductive layer is patterned using a screen printing process.

[0054] Based on the deformation region distribution of the target continuum robot body, a piezoresistive sensing layer is attached to the position corresponding to the high modulus sensing region on the second flexible substrate layer to cover the preset intersection area of ​​the second stretchable wire layer.

[0055] The first base layer printed with the first stretchable wire layer is laminated and assembled with the second flexible base layer having the piezoresistive sensing layer and the second stretchable wire layer, so that the first stretchable wire layer and the second stretchable wire layer intersect at a preset position, and the piezoresistive sensing layer is located at the intersection point.

[0056] The laminated structure is thermo-pressed to form a flexible array pressure sensor with heterogeneous modulus partitioning.

[0057] Compared with the prior art, the beneficial effects of the present invention are:

[0058] This invention provides an adaptive control method and system for a continuum robot based on global tactile perception, enabling direct force sensing across the entire robot body: By using a conformally attached flexible array pressure sensor, the contact force across the entire robot body is directly measured, overcoming the blind zone problem of existing end-effector force sensors. It also avoids the high cost and increased stiffness issues of fiber optic grating solutions, achieving low-cost, highly flexible global force sensing. Furthermore, it ensures measurement accuracy under deformation: the heterogeneous modulus structure design effectively solves the problem of measurement inaccuracies by flexible sensors under large deformation environments. It enhances interactive intelligence and safety: the adaptive admittance control strategy enables the robot to dynamically adjust its compliance characteristics based on real-time tactile information, significantly improving operational safety in sensitive environments (such as surgery). Finally, it enhances system compatibility: the sensor scheme and control framework have good universality and can be applied to continuum robot platforms with different structures.

[0059] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it according to the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Specific embodiments of the present invention are given in detail below with reference to the accompanying drawings. Attached Figure Description

[0060] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0061] Figure 1 A schematic diagram of the overall structure of a continuum robot for attaching flexible array pressure sensors;

[0062] Figure 2 This is a diagram showing the layered structure of a flexible array pressure sensor.

[0063] Figure 3 Example diagrams showing sensor units attached to different parts of a continuum robot;

[0064] Figure 4This is a diagram illustrating the overall architecture of an adaptive control system for a continuum robot based on global tactile perception.

[0065] Figure 5 The flowchart shows an adaptive control method for a continuum robot based on global tactile perception.

[0066] Figure 6 Flowchart for mapping two-dimensional pressure distribution data to a three-dimensional operating space;

[0067] Figure 7 This is a schematic diagram of the coordinate mapping relationship of a continuum robot;

[0068] Figure 8 Flowchart for adjusting adaptive admittance control parameters;

[0069] Figure 9 A flowchart for dynamically adjusting the smoothing parameters used for admittance control based on the determined contact force magnitude and distribution pattern;

[0070] Figure 10 This is a schematic diagram of a computer device.

[0071] Figure 11 This is a schematic diagram of a computer-readable storage medium. Detailed Implementation

[0072] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. It should be noted that, without conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.

[0073] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0074] The drawing numbers in this application are only used to distinguish the steps in the scheme and are not used to limit the execution order of the steps. The specific execution order is as described in the specification.

[0075] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0076] Example 1

[0077] To achieve full-domain tactile perception and intelligent interactive control of a continuum robot, this embodiment provides an adaptive control system for a continuum robot based on full-domain tactile perception, such as... Figure 4 As shown, it includes:

[0078] The flexible continuum robot body employs a multi-segment continuous bending structure to continuously deform in three-dimensional space, meaning it possesses the ability to continuously deform in three-dimensional space. Its driving methods include, but are not limited to, cable-driven, pneumatic-driven, or shape memory alloy-driven mechanisms.

[0079] A flexible array of pressure sensors is attached to the outer surface of the flexible continuum robot body to measure the pressure distribution on the outer surface in real time. Figure 1 , Figure 3 As shown.

[0080] Specifically, the flexible array pressure sensor adopts a heterogeneous modulus layered structure, such as... Figure 1 As shown, it includes a high-modulus sensing region and a low-modulus stretching region. The high-modulus sensing region is attached to the rigid section of the robot body, and the low-modulus stretching region is attached to the flexible section of the robot body.

[0081] The signal processing and control module is electrically connected to the actuators of the flexible array pressure sensor and the flexible continuum robot body.

[0082] The signal processing and control module is configured to: receive pressure distribution data from the flexible array pressure sensor; map the two-dimensional pressure distribution data output by the flexible array pressure sensor to a three-dimensional operating space based on the kinematic model of the robot body to determine the three-dimensional position of the contact force; dynamically adjust the smoothing parameters used for admittance control according to the determined contact force magnitude and distribution pattern; and generate motion control commands for the continuum robot body using the adjusted smoothing parameters to achieve compliant interaction.

[0083] The flexible continuum robot body, flexible array pressure sensors, and signal processing and control modules work together to form a closed loop of perception-decision-control.

[0084] like Figure 2 As shown, the heterogeneous modulus layered structure of the flexible array pressure sensor includes, from top to bottom:

[0085] First flexible substrate layer 1;

[0086] The first stretchable conductor layer 2 is formed of liquid metal material;

[0087] Piezoresistive sensing layer 3;

[0088] The second stretchable wire layer 4 is formed of liquid metal material and is arranged intersecting with the first stretchable wire layer to define a sensing unit array; that is, the first and second stretchable wire layers are arranged in a horizontal and vertical crisscross pattern, and each intersection point is a sensing unit, such as... Figure 2 As shown, an M×N sensor unit array is formed.

[0089] Second flexible substrate layer 5;

[0090] The piezoresistive sensing layer covers the intersection of the first stretchable wire layer and the second stretchable wire layer at the position corresponding to the high modulus sensing region, but does not cover the intersection at the position corresponding to the low modulus stretchable region.

[0091] The first and second flexible substrate layers utilize a modified thermoplastic polyurethane elastomer (TPU) film, which exhibits excellent tensile properties, with a Young's modulus reaching 10 MPa, and a film thickness of 50 μm. The liquid metal material is a gallium-indium alloy, and wires made from this material possess excellent stretchability and tensile strength. The piezoresistive sensing layer is made of Velostat material, which changes resistance under pressure.

[0092] The designed flexible array pressure sensor is attached segmentally along the length of the continuous robot body, covering its entire outer surface (i.e., the entire potential contact area). The attachment method is as follows: Figure 3 As shown, the high-modulus sensing area of ​​the flexible array pressure sensor is precisely attached to the rigid section or the non-deformation area with small deformation of the continuous robot body, and the low-modulus tensile area is attached to the flexible section (i.e., the deformation area) of the continuous robot body with the expected large deformation. The sensing unit of the sensor can be designed into the specific shape of the continuous robot and customized according to the geometry of the robot's outer surface.

[0093] The high-modulus sensing zone, located in the rigid section, utilizes piezoresistive material to ensure accurate pressure measurement. The low-modulus tensile zone, located in the deformation section, employs highly elastic TPU material to withstand large strain deformation. The liquid metal wire maintains stable conductivity under deformation, with an elongation rate exceeding 50%. Real-time measurement of the two-dimensional pressure distribution in the attachment area outputs a pressure distribution image / matrix. When the continuous robot undergoes complex deformations such as bending or torsion, the strain in its deformation zone is primarily absorbed by the sensor's low-modulus tensile zone, with minimal impact on the high-modulus sensing zone. Simultaneously, the liquid metal wire allows for free extension under deformation, ensuring the sensing unit's signal transmission function operates normally in any robot posture, achieving continuous and stable measurement of contact force.

[0094] In some embodiments, the hardware of the signal processing and control module includes: a signal conditioning circuit based on the AD8421 instrumentation amplifier to amplify and filter the signal; a data acquisition unit employing a multiplexer and an ADS131M08 ADC chip; and a communication interface using an RS485 bus for data upload. The software algorithm includes: kinematic modeling of the continuum robot, which maps the acquired two-dimensional pressure information distribution to three-dimensional space to determine the specific three-dimensional location of the contact on the robot body; and an adaptive admittance control algorithm that dynamically adjusts control parameters based on tactile information.

[0095] In some embodiments, the fabrication of the flexible array pressure sensor includes the following steps:

[0096] Step 1, Template Design and Preparation: Design a wire pattern template with a serpentine skeleton structure;

[0097] Two screen printing plates are prepared. The first plate is used to print the pattern of longitudinal conductors (e.g., 4 lines) (corresponding to the first stretchable conductor layer 2 on the first flexible substrate layer 1), and the second plate is used to print the pattern of transverse conductors (e.g., 21 lines) (corresponding to the second stretchable conductor layer 4 on the second flexible substrate layer 5).

[0098] 300-mesh polymer mesh and thermoplastic elastomer (SBS) material were selected as the temporary load-bearing substrate.

[0099] A modified thermoplastic polyurethane elastomer film is prepared as the first flexible substrate layer 1 and the second flexible substrate layer 5, and is laid flat on the SBS substrate.

[0100] Step 2, Conductive Layer Patterning (Liquid Metal Conductors): Stir the gallium-indium alloy liquid metal thoroughly (approximately 10 minutes) before use to ensure uniform mixing;

[0101] The vertical and horizontal guide patterns are printed sequentially using screen printing technology;

[0102] After printing, dry at 80°C for 10 minutes to remove solvent and pre-cure;

[0103] The printed conductors are lightly stretched to activate them and their conductivity and stability are checked.

[0104] Step 3, Sensor Layer Integration and Heterogeneous Modulus Structure Implementation: High Modulus Sensing Region Location: Based on the structure of the target continuum robot (distribution of rigid segments / deformation regions), determine the location where high modulus sensing units need to be set on the sensor.

[0105] Sensing layer bonding: Velostat material (or equivalent piezoresistive material) is precisely cut and bonded to the predetermined high modulus sensing area of ​​the second stretchable conductor layer 4 on the second flexible substrate layer 5 (i.e., covering the intersection area of ​​the transverse conductor and the subsequent longitudinal conductor).

[0106] Lamination assembly: The first flexible substrate layer 1 (with the conductor face down) printed with the first stretchable conductor layer 2 is precisely aligned and stacked with the second flexible substrate layer 5 (with the conductor face up) with the sensing layer and the second stretchable conductor layer 4, ensuring that the longitudinal conductor and the transverse conductor intersect at a predetermined position and that the sensing layer is located at the intersection point.

[0107] Low-modulus stretching zone retention: In non-sensing areas (i.e., locations designed as low-modulus stretching zones), only the TPU base layer (without Velostat material) is retained to ensure high elasticity in this area.

[0108] Step 4, Encapsulation and Curing: Place the assembled multilayer sensor structure into a hot press;

[0109] Hot pressing at 150°C for 15 seconds causes the TPU substrate to melt slightly, achieving strong interlayer bonding and overall encapsulation while maintaining flexibility.

[0110] The high-modulus sensing area is precisely aligned with the rigid segment of the continuum robot body, while the low-modulus tensile area strictly corresponds to the flexible segment (the deformation zone of the continuum robot) where large deformation is expected. A segmented attachment strategy is adopted to cover the entire outer surface of the robot.

[0111] The flexible array pressure sensor is connected to the signal processing and control module via a stretchable wire. The signal conditioning circuit (instrumentation amplifier AD8421) in this module amplifies and filters the original signal, the multiplexer realizes time-division multiplexing acquisition of the array signal, and the ADC (ADS131M08) chip converts the analog signal into a digital signal. The microcontroller processes and packages the data, and finally the data is uploaded to the host computer via the RS485 bus.

[0112] For a detailed description of the adaptive control method for the continuum robot used in this system, please refer to the corresponding description in the following method embodiments, which will not be repeated here.

[0113] Example 2

[0114] An adaptive control method for a continuum robot based on full-domain tactile perception is presented. Based on the aforementioned system, a detailed description of the system can be found in the corresponding description in the system embodiments described above, and will not be repeated here. A flexible array of pressure sensors is attached to the outer surface of the continuum robot. These flexible array pressure sensors are sensors with heterogeneous modulus partitions; the high-modulus sensing region is attached to the rigid section of the continuum robot, and the low-modulus tensile region is attached to the deformable section of the continuum robot. This ensures the continuity and stability of the pressure measurement signal under all postures of the continuum robot. Figure 5 As shown, the method includes the following steps:

[0115] S100: Obtain the two-dimensional pressure distribution data output by the flexible array pressure sensor;

[0116] S200. Based on the kinematic model of the continuum robot, the two-dimensional pressure distribution data is mapped to the three-dimensional operating space to determine the three-dimensional position of the contact force.

[0117] To achieve the spatialization and manipulability of tactile information and overcome the core challenges posed by the deformation of the continuum robot body, such as Figure 6 As shown, the steps for mapping the two-dimensional pressure distribution data to the three-dimensional operating space include:

[0118] S210. For each sensing unit that outputs a pressure signal, determine its position in the local coordinate system of the continuum robot body;

[0119] S220. Based on the current joint state of the continuum robot, calculate the homogeneous transformation matrix from the base coordinate system of the continuum robot to the local coordinate system where the sensing unit is located.

[0120] This embodiment establishes the mapping relationship from joint space to operating space and from drive space to joint space, as shown in the coordinate mapping example. Figure 7 As shown. The mapping from joint space to operand space can be achieved through a homogeneous transformation matrix from the base coordinate system to the end effector coordinate system. Implementation. This transformation process includes starting from the center point of the base coordinate system. to the center point of the end coordinate system Translation, and sequential rotation around Axis rotation Angle, around Axis rotation Corner, then around Axis rotation The angle is used to obtain the homogeneous transformation matrix as shown in formula (1). In the formula, Represents the sine function , Represents the cosine function ,That middle, , Let be the length of the continuous robot. Using formula (1), the bending angle can be determined... and rotation angle The coordinates of the end effector of the continuum robot in the operating space are obtained by solving.

[0121]

[0122] S230. Through spatial mapping, the three-dimensional position of the sensing unit in the world coordinate system is calculated, thereby associating each pressure signal with a three-dimensional spatial contact point. The formula is as follows:

[0123]

[0124] in, Let be the homogeneous transformation matrix from the base to the end. The position of the sensing unit in the local coordinate system. This represents the position of the sensing unit in the Cartesian coordinate system.

[0125] From formula (1), we can obtain the joint space variable expression for the end position of the continuum robot, namely formula (3).

[0126]

[0127] This embodiment is based on the established kinematic model of a continuum robot, and achieves a precise mapping from the two-dimensional coordinates of the sensor to the three-dimensional operating space of the robot through a homogeneous transformation matrix. For each sensing unit, its position in the world coordinate system is determined by forward kinematics calculation (Equation (2)). Thus, each force signal in the two-dimensional pressure distribution map of the sensor is precisely associated with the specific contact position in the three-dimensional operating space.

[0128] This embodiment transforms tactile information from an abstract data matrix into a three-dimensional spatial vector that can be directly understood and used by motion control algorithms, serving as the spatial benchmark for all subsequent intelligent decisions.

[0129] Because the shape of a continuum robot changes constantly, the absolute position of the sensors in space shifts significantly, rendering traditional fixed calibration methods completely ineffective. This embodiment dynamically binds the local coordinates of the sensing units to the robot's current real-time posture by calculating a homogeneous transformation matrix in real time. Regardless of the robot's twisted shape, the system can instantly calculate the precise position of each sensing unit in the three-dimensional world, achieving posture-independent, all-time, all-pose tactile localization.

[0130] S300. Based on the determined magnitude and distribution pattern of the contact force, dynamically adjust the smoothing parameters used for admittance control; wherein, the smoothing parameters include the desired virtual stiffness value and the desired virtual damping value;

[0131] The adjustment strategy for the smoothing parameters used in admittance control is as follows: the greater the contact force, the smaller the desired virtual stiffness value, and the greater the desired virtual damping value.

[0132] To resolve the precision-compliance trade-off in interactive control and to achieve a combination of active safety and passive compliance, such as Figure 8 , Figure 9 As shown, the step of dynamically adjusting the smoothing parameters used for admittance control based on the determined contact force magnitude and distribution pattern includes:

[0133] S310. Set a pressure threshold and a safety threshold, wherein the safety threshold is greater than the pressure threshold; for example, set the pressure threshold to 0.3N and the safety threshold to 1N.

[0134] S320. When the pressure of all sensing units is less than the pressure threshold, it is determined to be a non-contact state. The desired virtual stiffness value is adjusted to the first virtual stiffness range, and the desired virtual damping value is adjusted to the first virtual damping range to ensure positioning accuracy and system response. For example, the initial value of virtual stiffness is set to 20, and the initial value of virtual damping is set to 10. When in a non-contact state, the desired virtual stiffness value is maintained at 90% to 100% of the initial value of virtual stiffness (i.e., 18 to 20), and the desired virtual damping value is maintained at 90% to 110% of the initial value of virtual damping (i.e., 9 to 11) to ensure positioning accuracy and system response.

[0135] S330. When the pressure of any sensing unit is greater than the pressure threshold and the pressure of all sensing units is less than the safety threshold, a slight contact state is determined. The desired virtual stiffness value is adjusted to the second virtual stiffness range, and the desired virtual damping value is adjusted to the second virtual damping range to enhance the robot's compliance and interaction smoothness. The second virtual stiffness range is smaller than the first virtual stiffness range, and the second virtual damping range is larger than the first virtual damping range. For example, when entering a slight contact state (where the local contact force is between the set pressure threshold and the safety threshold, such as 0.3N to 1N), the desired virtual stiffness value is smoothly reduced to 60% to 80% of the initial virtual stiffness value (i.e., 12 to 16), and the desired virtual damping value is moderately increased to 120% to 150% of the initial virtual damping value (i.e., 12 to 15) to enhance the robot's compliance and interaction smoothness.

[0136] S340. When the pressure of any sensing unit exceeds the safety threshold, a collision risk state is determined. The desired virtual stiffness value is adjusted to the third virtual stiffness range, and the desired virtual damping value is simultaneously adjusted to the third virtual damping range to achieve rapid impact absorption and active avoidance. The third virtual stiffness range is smaller than the second virtual stiffness range, and the third virtual damping range is larger than the second virtual damping range. For example, when a collision risk is detected (when a local high-pressure point exceeds the safety threshold, such as greater than 1N), the desired virtual stiffness value is significantly reduced to 20%–40% of the initial virtual stiffness value (i.e., 4–8), and the desired virtual damping value is significantly increased to 200%–300% of the initial virtual damping value (i.e., 20–30) to achieve rapid impact absorption and active avoidance.

[0137] For admittance controllers with fixed parameters, high stiffness ensures trajectory tracking accuracy but generates a huge impact force upon contact. Low stiffness can compliantly absorb the impact but sacrifices positioning accuracy. This embodiment dynamically adjusts the stiffness and damping values ​​of the robot during periods of no or minimal contact (small contact force). In this case, the robot can execute predetermined free-space movements with high precision and high response speed, ensuring operational accuracy. During periods of strong contact / collision (large contact force), low stiffness and high damping values ​​are used. In this case, the robot actively reduces its stiffness to conform to the external force while increasing damping to quickly dissipate kinetic energy, thereby maximizing impact buffering and absorbing collision energy to protect itself and the environment. This is not only passive buffering (passive compliance) after a collision but also an active safety strategy based on tactile prediction. When the sensor detects a sudden increase in local pressure (indicating a collision risk), the controller reduces stiffness in advance before a substantial hard collision occurs, changing hard-on-hard contact to soft contact and minimizing potential damage.

[0138] S400. Using the adjusted smoothing parameters, generate motion control commands for the continuum robot body to achieve compliant interaction.

[0139] When a continuum robot comes into contact with its environment, the interaction forces cause deformation in both. To prevent damage to the robot or environmental objects, an admittance control model is introduced to describe the contact relationship, and each contact point of the robot body is equivalent to a second-order mass-spring-damped system. First, the admittance model of a single-point force source is analyzed, and its expected admittance mathematical model is shown in formula (4).

[0140]

[0141] In the formula, This refers to the actual interaction force between the robot and the environment, measured by a pressure sensor. These represent the inertia, damping, and stiffness parameters of the admittance model, respectively. These represent the robot's desired acceleration, velocity, and position, respectively. These represent the actual acceleration, velocity, and position, respectively. The position deviation is defined as... Substituting into the above formula, we get formula (5).

[0142]

[0143] in, For virtual inertial mass, This is the position correction amount. For speed correction, For acceleration correction, This is the current contact force vector.

[0144] Taking a Laplace transform of both sides of equation (5), we can obtain the transfer function of admittance control as follows:

[0145]

[0146] From formula (6), the pole expression of the admittance control system model can be obtained:

[0147]

[0148] According to the system stability criterion, when At that time, the admittance control model is stable.

[0149] To enable continuum robots to adapt to complex interactive tasks, this invention provides an adaptive admittance control strategy. The core of this strategy lies in dynamically adjusting the parameters of the admittance controller, i.e., the virtual stiffness, based on real-time information fed back from a flexible array pressure sensor. and virtual damping .

[0150] The core idea of ​​the parameter adaptive control method is to construct a controller based on adaptive admittance, i.e., the basic admittance model, based on the magnitude of the contact force and formula (5), with the inertia coefficient initially set. Damping coefficient rigidity parameters To ensure a fast and stable system response, the virtual stiffness is smoothly adjusted based on the total contact force. and virtual damping .

[0151] The virtual stiffness is dynamically adjusted based on the magnitude and distribution pattern of the contact force, i.e., no contact, slight contact, and collision risk. and virtual damping Specifically, when there is no contact, the pressure of all sensing units is less than a set threshold. Maintain high value, Maintain the median value to ensure control accuracy; when there is slight contact, the local force should be less than the set threshold and safety value. Smooth decrease, A slight increase, initiating compliant control; when there is a risk of collision, localized high-pressure points exist. Significantly reduced, The number of cases has increased significantly, and people are starting to actively avoid them.

[0152] During the adjustment of smoothing parameters, ensure that the system stability conditions are always met: The adaptive admittance control parameter adjustment process is as follows: Figure 8 , Figure 9 As shown. To avoid system instability caused by abrupt changes in admittance parameters, after the step of dynamically adjusting the smoothing parameters used for admittance control based on the determined contact force magnitude and distribution pattern, the following is also included:

[0153] The desired virtual stiffness value and the desired virtual damping value are smoothed to obtain the smoothed virtual stiffness value and smoothed virtual damping value actually used in the current control cycle.

[0154] In this embodiment, all admittance parameter adjustments are smoothly transitioned using a first-order low-pass filter, and the smooth transition formula is as follows:

[0155]

[0156] Among them, among them, and The virtual stiffness and virtual damping at the current moment, and For the desired virtual stiffness and desired virtual damping determined, and These are the virtual stiffness and virtual damping values ​​from the previous control cycle. It is a smoothing factor, and Its value determines the rate at which the parameters are updated. The smaller the value, the smoother the transition; The larger the value, the faster the response, making it suitable for dynamic interactive scenarios.

[0157] To achieve precise injection of dynamic compliance characteristics and ensure progressive stability and smooth response of the control system, the step of generating motion control commands for the continuum robot body using adjusted smoothing parameters to achieve compliant interaction includes:

[0158] Substituting the smoothed virtual stiffness value, the smoothed virtual damping value, and the current contact force into the admittance model, the motion correction amount of the continuum robot is calculated, and the final control command is generated accordingly to drive the continuum robot to achieve compliant interaction with the environment.

[0159] The admittance model itself defines the robot's response characteristics to external forces. By substituting smoothed parameters and real-time measured contact forces into the model, the system is actually reconstructing the robot's dynamic equations in real time. This makes the robot's dynamic behavior at the moment of contact resemble a virtual spring-damped system with specific stiffness and damping. This dynamic change in virtual characteristics is the direct physical reason why the robot exhibits intelligent compliance (sometimes stiff, sometimes soft).

[0160] Using smoothed virtual stiffness and damping values, filtered by a first-order low-pass filter, instead of directly abruptly changing expected values, avoids command shocks. The smooth changes in parameters ensure that the motion corrections and their derivatives (velocity and acceleration corrections) output by the admittance model are also continuous and smooth. This fundamentally eliminates step jumps or jitters in motor torque commands caused by abrupt parameter changes, protecting the drive system. The smooth transition gives the control system sufficient time to adapt to the new dynamic characteristics, avoiding instantaneous instability or oscillations that may be caused by drastic changes in model parameters, ensuring the robustness of the entire interaction process.

[0161] The joint changes of a tethered continuum robot are controlled by the length of the driving cable. To achieve better control, a mapping relationship from joint space to driving space needs to be established. The change in cable length controls the bending angle. and rotation angle During the bending process of a continuum robot, the centerline... and drive cables The bending angles are equal, but there is a positional offset between the drive cable and the center line, and the difference in the radius of curvature of the bends is equal to the radius of the continuum robot. Now assume the rotation angle When the length of the first drive cable changes, it is transformed into formula (9).

[0162]

[0163] When the bending angle of the continuum robot is The rotation angle is At that time, the change in length of the four equally spaced cables is Specifically, as shown in formula (10), this represents the mapping from joint space to drive space for the continuum robot. Based on geometric relationships, the length variation of each drive cable can be expressed as:

[0164]

[0165] The adaptive control method for a continuum robot based on global tactile perception provided by this invention can achieve direct force perception across the entire domain, ensure measurement accuracy under deformation, improve interactive intelligence and safety, and enhance system compatibility.

[0166] Example 3

[0167] A computer device 500, such as Figure 10 As shown, the system includes a memory 510, a processor 520, and a computer program 530 stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of an adaptive control method for a continuum robot based on global tactile perception. For a detailed description of the method, please refer to the corresponding description in the above method embodiments; it will not be repeated here.

[0168] Example 4

[0169] A computer-readable storage medium, such as Figure 11 As shown, a computer program is stored thereon. When executed by a processor, the computer program implements the steps of an adaptive control method for a continuum robot based on global tactile perception. For a detailed description of the method, please refer to the corresponding description in the above method embodiments, which will not be repeated here.

[0170] The number of devices and processing scale described herein are for the purpose of simplifying the description of the invention. Applications, modifications, and variations of the invention will be readily apparent to those skilled in the art.

[0171] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

[0172] The apparatus, computer device, and non-volatile computer storage medium and method provided in the embodiments of this specification are corresponding. Therefore, the apparatus, computer device, and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding apparatus, computer device, and non-volatile computer storage medium will not be repeated here.

[0173] Those skilled in the art will also know that, besides implementing the controller in the form of purely computer-readable program code, the same functions can be achieved by logically programming the method steps, making the controller take the form of logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers (PLCs), and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the devices included within it for implementing various functions can also be considered structures within that hardware component. Alternatively, the devices for implementing various functions can be considered as both software units implementing the method and structures within a hardware component.

[0174] The systems, apparatuses, or units described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. For ease of description, the above apparatuses are described separately by function as various units. Of course, when implementing one or more embodiments of this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0175] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0176] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0177] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0178] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0179] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0180] This specification may be described in the general context of computer-executable instructions, such as program units, that are executed by a computer. Generally, program units include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification may also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program units may reside in local and remote computer storage media, including storage devices.

[0181] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0182] The above description is merely an embodiment of this specification and is not intended to limit the scope of one or more embodiments of this specification. Various modifications and variations can be made to one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of one or more embodiments of this specification.

Claims

1. An adaptive control method for a continuum robot based on global tactile perception, characterized in that, A flexible array pressure sensor is attached to the outer surface of the continuous robot. This flexible array pressure sensor has heterogeneous modulus partitions; its high-modulus sensing region is attached to the rigid section of the continuous robot, and its low-modulus tensile region is attached to the deformable section of the continuous robot. This ensures the continuity and stability of the pressure measurement signal under all postures of the continuous robot. The method includes the following steps: Acquire the two-dimensional pressure distribution data output by the flexible array pressure sensor; Based on the kinematic model of the continuum robot, the two-dimensional pressure distribution data is mapped to the three-dimensional operating space to determine the three-dimensional position of the contact force. Based on the determined contact force magnitude and distribution pattern, the smoothing parameters used for admittance control are dynamically adjusted. By using the adjusted smoothing parameters, motion control commands for the continuum robot body are generated to achieve compliant interaction; The smoothing parameters include the desired virtual stiffness value and the desired virtual damping value; The adjustment strategy for the smoothing parameter used in admittance control is as follows: the greater the contact force, the smaller the expected virtual stiffness value, and the greater the expected virtual damping value. The step of dynamically adjusting the smoothing parameters used for admittance control based on the determined contact force magnitude and distribution pattern includes: Set a pressure threshold and a safety threshold, wherein the safety threshold is greater than the pressure threshold; When the pressure of all sensing units is less than the pressure threshold, it is determined to be a non-contact state. The desired virtual stiffness value is adjusted to the first virtual stiffness range, and the desired virtual damping value is adjusted to the first virtual damping range to ensure positioning accuracy and system response. When the pressure of any sensing unit exceeds the pressure threshold and the pressure of all sensing units is less than the safety threshold, it is determined to be a slight contact state. The desired virtual stiffness value is adjusted to the second virtual stiffness range, and the desired virtual damping value is adjusted to the second virtual damping range to enhance the robot's compliance and interaction smoothness. The second virtual stiffness range is smaller than the first virtual stiffness range, and the second virtual damping range is larger than the first virtual damping range. When the pressure of any sensing unit exceeds the safety threshold, it is determined to be a collision risk state. The desired virtual stiffness value is adjusted to the third virtual stiffness range, and the desired virtual damping value is adjusted to the third virtual damping range to achieve rapid impact absorption and active avoidance. The third virtual stiffness range is smaller than the second virtual stiffness range, and the third virtual damping range is larger than the second virtual damping range.

2. The adaptive control method for a continuum robot based on global tactile perception as described in claim 1, characterized in that, The steps of mapping the two-dimensional pressure distribution data to the three-dimensional operating space include: For each sensing unit that outputs a pressure signal, determine its position in the local coordinate system of the continuum robot body; Based on the current joint state of the continuum robot, calculate the homogeneous transformation matrix from the base coordinate system of the continuum robot to the local coordinate system where the sensing unit is located, using the following formula: in, Represents the sine function , Represents the cosine function , Indicates the bending angle. , Indicates the rotation angle. , The length of the continuum robot; By using spatial mapping relationships, the three-dimensional position of the sensing unit in the Cartesian coordinate system is calculated, thereby associating each pressure signal with a three-dimensional spatial contact point. The formula is as follows: in, Let be the homogeneous transformation matrix from the base to the end. The position of the sensing unit in the local coordinate system. This represents the position of the sensing unit in the Cartesian coordinate system.

3. The adaptive control method for a continuum robot based on global tactile perception as described in claim 1, characterized in that, Following the step of dynamically adjusting the smoothing parameters used for admittance control based on the determined contact force magnitude and distribution pattern, the method further includes: The desired virtual stiffness value and the desired virtual damping value are smoothed to obtain the smoothed virtual stiffness value and smoothed virtual damping value actually used in the current control cycle.

4. The adaptive control method for a continuum robot based on global tactile perception as described in claim 3, characterized in that, The smoothing process for the desired virtual stiffness value and the desired virtual damping value is specifically achieved through a first-order low-pass filter, with the following formula: in, and These are the virtual stiffness and virtual damping values ​​at the current moment. and To determine the desired virtual stiffness and desired virtual damping values, and These are the virtual stiffness and virtual damping values ​​from the previous control cycle. It is a smoothing factor, and .

5. The adaptive control method for a continuum robot based on global tactile perception as described in claim 4, characterized in that, During the adjustment of smoothing parameters, ensure that the system stability conditions are always met: ,in The virtual inertial mass of the admittance controller.

6. The adaptive control method for a continuum robot based on global tactile perception as described in claim 3, characterized in that, The step of generating motion control commands for the continuum robot body using adjusted smoothing parameters to achieve compliant interaction includes: Substituting the smoothed virtual stiffness value, the smoothed virtual damping value, and the current contact force into the admittance model, the motion correction amount of the continuum robot is calculated, and the final control command is generated accordingly to drive the continuum robot to achieve compliant interaction with the environment.

7. The adaptive control method for a continuum robot based on global tactile perception as described in claim 6, characterized in that, The admittance model is as follows: in, For virtual inertial mass, This is the position correction amount. For speed correction, For acceleration correction, This is the current contact force vector.

8. An adaptive control system for a continuum robot based on global tactile perception, employing the method as described in any one of claims 1 to 7, characterized in that, include: The flexible continuum robot body adopts a multi-segment continuous bending structure to continuously deform in three-dimensional space; A flexible array pressure sensor is attached to the outer surface of the flexible continuum robot body to measure the pressure distribution on the outer surface in real time. The signal processing and control module is electrically connected to the actuators of the flexible array pressure sensor and the flexible continuum robot body. The flexible array pressure sensor adopts a heterogeneous modulus layered structure, which includes a high modulus sensing region and a low modulus tensile region. The high modulus sensing region is attached to the rigid section of the flexible continuum robot body, and the low modulus tensile region is attached to the flexible section of the flexible continuum robot body. The signal processing and control module is configured to: receive two-dimensional pressure distribution data from the flexible array pressure sensor; map the two-dimensional pressure distribution data output by the flexible array pressure sensor to a three-dimensional operating space based on the kinematic model of the flexible continuum robot body to determine the three-dimensional position of the contact force; dynamically adjust the smoothing parameters used for admittance control according to the determined contact force magnitude and distribution pattern; and generate motion control commands for the flexible continuum robot body using the adjusted smoothing parameters to achieve compliant interaction.

9. The adaptive control system for a continuum robot based on global tactile perception as described in claim 8, characterized in that, The heterogeneous modulus layered structure of the flexible array pressure sensor includes, from top to bottom: First flexible substrate layer; The first stretchable conductor layer is formed of liquid metal material; Piezoresistive sensing layer; The second stretchable wire layer is formed of liquid metal material and is arranged to intersect with the first stretchable wire layer to define the sensing unit array. Second flexible substrate layer; The piezoresistive sensing layer covers the intersection of the first stretchable wire layer and the second stretchable wire layer at the position corresponding to the high modulus sensing region, but does not cover the intersection at the position corresponding to the low modulus stretchable region.

10. The adaptive control system for a continuum robot based on global tactile perception as described in claim 9, characterized in that, The first flexible substrate layer and the second flexible substrate layer are made of thermoplastic polyurethane elastomer; the liquid metal material is gallium indium alloy; and the piezoresistive sensing layer is made of Velostat material.

11. The adaptive control system for a continuum robot based on global tactile perception as described in claim 9, characterized in that, The fabrication of the flexible array pressure sensor includes the following steps: Provide a first flexible substrate layer and a second flexible substrate layer; On the first flexible substrate layer, a first stretchable conductive layer is patterned using a screen printing process; on the second flexible substrate layer, a second stretchable conductive layer is patterned using a screen printing process. Based on the deformation region distribution of the target continuum robot body, a piezoresistive sensing layer is attached to the position corresponding to the high modulus sensing region on the second flexible substrate layer to cover the preset intersection area of ​​the second stretchable wire layer. The first base layer printed with the first stretchable wire layer is laminated and assembled with the second flexible base layer having the piezoresistive sensing layer and the second stretchable wire layer, so that the first stretchable wire layer and the second stretchable wire layer intersect at a preset position, and the piezoresistive sensing layer is located at the intersection point. The laminated structure is thermo-pressed to form a flexible array pressure sensor with heterogeneous modulus partitioning.

Images