Sensing skin and sensing method for protective robot

Abstract

The invention discloses a sensing skin and a sensing method for protecting a robot. The sensing skin comprises: a sensing skin wrapped around the outside of a mechanical arm and a controller electrically connected to the sensing skin. The sensing method applies an alternating electric field with a frequency of 10-100kHz and a voltage of 5-30V to an electrostatic sensing unit array on a flexible base layer of the mechanical arm, and collects capacitance change signals of the electrostatic sensing units. The capacitance change signals are amplified by a programmable gain amplifier, interference is eliminated by digital filtering, and digital sampling is performed by an analog-to-digital converter. Based on the processed capacitance change signals, the human body approach distance is calculated by an equivalent capacitance model and a danger level assessment model is established, and a warning signal is issued based on the assessment result. The invention can effectively improve the safety performance of industrial robots, provide reliable safety protection for human-machine collaboration, and has broad application prospects.

Description

Technical Field

[0001] The present invention belongs to the technical field of robot sensing skin, and in particular relates to sensing skin and a sensing method for protecting a robot. Background Art

[0002] In the actual production process, workers are often required to operate or maintain the robot within the working range, which puts higher requirements on the safety protection of the robot. Traditional safety protection methods mainly use hard isolation methods such as safety fences and photoelectric sensors. Although they can achieve basic safety protection functions, they severely limit the flexibility of human-machine collaboration.

[0003] To solve the above problems, the industry has proposed a variety of flexible protection solutions based on sensors, such as using a combination of infrared ranging sensor arrays and pressure sensors to achieve safety protection. However, infrared sensors are easily interfered by ambient light and may cause misjudgment under strong light conditions; pressure sensors can only be triggered after actual contact occurs and cannot provide early warning; in addition, the solution does not consider dynamic factors such as human approach speed and direction, and the safety assessment is not comprehensive enough. Summary of the invention

[0004] The present invention provides a sensing skin and a sensing method for protecting a robot. Based on a flexible base layer and a partitioned protective layer design, a danger level assessment model based on distance, angle, and speed is established, and a three-level safety warning mechanism is provided. The robot can be adjusted in real time and emergency stopped, thus forming a complete robot safety protection solution.

[0005] The specific plan is as follows:

[0006] In a first aspect, the present invention provides a sensing skin for protecting a robot, wherein the sensing skin comprises: a sensing skin wrapped around the outside of a robot arm and a controller electrically connected to the sensing skin.

[0007] The sensing skin comprises, from the inside to the outside, a flexible base layer arranged on the robotic arm, an array of electrostatic induction units evenly distributed on the surface of the flexible base layer, and a partitioned protective layer for covering and protecting the induction units.

[0008] The controller includes: an electrostatic field excitation unit and a signal conditioning unit. The electrostatic field excitation unit is used to apply an alternating electric field to the electrostatic induction unit array, with an operating frequency of 10-100kHz and an excitation voltage of 5-30V; the signal conditioning unit is used to receive the capacitance change signal of the electrostatic induction unit, and the signal conditioning unit includes a signal amplification module, a digital filtering module and an analog-to-digital conversion module.

[0009] Furthermore, the flexible base layer includes a flexible circuit board made of polyimide material with a thickness of 50-200 microns; an interconnected wire network with a line width of 50-100 microns is provided on the surface of the flexible circuit board for connecting the electrostatic induction unit; a conductive shielding layer is provided at the bottom of the flexible circuit board for suppressing electromagnetic interference from the robotic arm body.

[0010] Furthermore, the electrostatic induction unit array includes metal electrodes, which are made of copper, aluminum or silver conductive materials with a thickness of 1-5 microns and an electrode size of 2 mm×2 mm to 5 mm×5 mm; the electrode spacing is 0.1-0.5 mm.

[0011] Furthermore, the partitioned protective layer includes: a plurality of independent silicone blocks with a thickness of 0.5-2 mm; gaps are provided between adjacent silicone blocks, with a gap width of 0.5-1.5 mm; and the gaps are filled with conductive fabrics for electrostatic shielding.

[0012] Furthermore, the controller also includes: a data processing unit for analyzing the approach distance and approach speed of a human body; and an early warning control unit for outputting corresponding control instructions according to the danger level.

[0013] In a second aspect, based on the sensing skin of the first aspect, the present invention also provides a sensing method for a protective robot, the sensing method comprising the following steps:

[0014] Step S1, applying an alternating electric field with a frequency of 10-100 kHz and a voltage of 5-30 V to the electrostatic induction unit array on the flexible base of the robot arm, and collecting the capacitance change signal of the electrostatic induction unit.

[0015] Step S2, amplifying the capacitance change signal using a programmable gain amplifier, eliminating interference through digital filtering, performing digital sampling through an analog-to-digital converter, and calculating the human body approach distance through an equivalent capacitance model based on the processed capacitance change signal.

[0016] Step S3, establishing a danger level assessment model according to the human body approach distance, the angle between the human body approach direction and the robot arm movement direction, and the human body approach speed, and issuing a warning signal based on the assessment result.

[0017] Further, based on the processed capacitance change signal Calculate the human body approach distance The equivalent capacitance model is: ,in, is the distance correction coefficient, the value range is 0.8-1.2, is the dielectric constant of vacuum, and A is the area of ​​the sensing unit array where the capacitance signal changes.

[0018] Furthermore, the risk level assessment model is: , where R is the risk level index, and its value range is 0 - 100. is the human body approaching distance, with the unit of meter. is the angle between the human body approaching direction and the robotic arm movement direction, with the unit of radian. is the human body approaching speed, with the unit of m / s. The weight coefficients , , are obtained as follows:

[0019] , , , where , are the maximum and minimum values of the effective detection distance, taking 0.2 m and 0.05 m respectively. is the approaching speed reference value, taking 10 m / s.

[0020] Furthermore, according to the magnitude of the risk level index R, the risk level is divided into:

[0021] Safe state: R ≤ 30, the robotic arm operates normally.

[0022] Warning state: 30 < R ≤ 70, reduce the robotic arm operation speed to 50%.

[0023] Dangerous state: R > 70, the robotic arm stops urgently.

[0024] Furthermore, the correction factor is calculated through the multi-point collaborative perception algorithm to correct the human body approaching distance d. The expression of the correction factor is:

[0025]

[0026] where ΔCi is the capacitance change value of the i-th adjacent sensing unit, ΔCmax is the maximum capacitance change value within the current detection period, k2 is the dynamic response coefficient, and its value range is 0.1 - 0.3, dv / dt is the change rate of the approaching distance, with the unit of m / s.

[0027] Furthermore, the corrected human body approaching distance .

[0028] Compared with the prior art, the beneficial effects of the present invention are:

[0029] The present invention wraps a flexible sensing skin on the outside of the robotic arm and adopts the principle of electrostatic induction to detect the approach distance, direction and speed of the human body in real time, providing all-round safety protection; based on the hazard level assessment model, combined with multiple parameters such as the approach distance, approach angle and speed of the human body, it realizes an accurate assessment of the dangerous situation and can automatically execute corresponding safety control measures (normal operation, reduced speed operation or emergency stop); the present invention can effectively improve the safety performance of industrial robots, provide reliable safety guarantees for human-machine collaboration, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a schematic diagram of the composition of the sensing skin for protecting a robot according to the present invention;

[0031] Figure 2 This is a flow chart of the sensing method for protecting a robot according to the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] Example 1

[0034] like Figure 1 As shown, it is a schematic diagram of the composition of the sensing skin for protecting the robot provided by the present invention, wherein the sensing skin comprises: a sensing skin wrapped around the outside of the robot arm and a controller electrically connected to the sensing skin.

[0035] The sensing skin comprises, from the inside to the outside, a flexible base layer arranged on the robotic arm, an array of electrostatic induction units evenly distributed on the surface of the flexible base layer, and a partitioned protective layer for covering and protecting the induction units.

[0036] The sensing skin of the present invention adopts a fully wrapped design, which can completely cover the working range of the robot arm. Taking a 6-axis industrial robot as an example, the sensing skin needs to wrap the outer surface of the 1st to 6th axes of the robot arm, especially adopting a segmented design at the joint position to ensure that the flexibility of the robot arm is not affected during movement. In addition, the controller adopts a distributed layout, and the signal acquisition front end is installed near the sensing unit to reduce the signal transmission distance and improve the signal quality.

[0037] The electrostatic induction unit array includes a metal electrode, which is made of copper, aluminum or silver conductive material with a thickness of 1-5 microns, an electrode size of 2 mm × 2 mm to 5 mm × 5 mm; the electrode spacing is 0.1-0.5 mm. The specific design of the metal electrode needs to balance the detection sensitivity and spatial resolution. For example, in practical applications, a 3mm × 3mm copper electrode with a thickness of 2 microns and an electrode spacing of 0.2 mm can be selected. This configuration can achieve a millimeter-level position resolution within a distance of 15 cm. The electrodes are made using a photolithography process to ensure consistency in size and spacing. In addition, the electrode surface needs to be treated for oxidation resistance, and a gold plating process with a thickness of 0.1 microns can be used to improve conductivity and service life.

[0038] The partitioned protective layer includes: multiple independent silicone blocks with a thickness of 0.5-2 mm; gaps are set between adjacent silicone blocks with a width of 0.5-1.5 mm; the gaps are filled with conductive fabric for electrostatic shielding. The innovative design of the partitioned protective layer greatly improves the practicality of the system. The silicone blocks are made of medical-grade silicone material with excellent biocompatibility and wear resistance. The area of ​​each silicone block is 10mm×10mm, and the conductive fabric filled in the gaps between adjacent blocks is made of silver fiber material, which not only has a good shielding effect, but also has a certain elasticity and can adapt to the movement and deformation of the robotic arm. In the experiment, it was found that this design can increase the strength of the induction signal by more than 30%.

[0039] The flexible base layer includes a flexible circuit board made of polyimide material with a thickness of 50-200 microns; an interconnection wire network with a line width of 50-100 microns is provided on the surface of the flexible circuit board for connecting the electrostatic induction unit; a conductive shielding layer is provided at the bottom of the flexible circuit board for suppressing electromagnetic interference from the robot body. The design of the flexible base layer fully considers the balance between mechanical and electrical properties. Polyimide material has excellent high temperature resistance (can remain stable in the range of -200°C to 300°C) and mechanical strength (tensile strength>100MPa). In practical applications, a flexible circuit board with a thickness of 100 microns can be selected, the interconnection wires on the surface have a line width of 75 microns, and the conductive shielding layer at the bottom uses a grounding copper foil with a thickness of 18 microns. This structure can work stably for more than 1 million times under the condition that the bending radius is greater than 5mm.

[0040] The three-layer structure design fully considers the actual application needs. Take a typical industrial robot work unit as an example. Assuming the length of the robot arm is 1.5 meters and the working radius is 2 meters, about 0.5 square meters of sensing skin is needed to cover its key parts. The flexible base layer as the bottom layer ensures good fit with the robot arm; the electrostatic induction unit array in the middle provides sensing function; the outermost partitioned protection layer protects the internal components while improving the sensitivity through a special partition design.

[0041] The controller includes: an electrostatic field excitation unit and a signal conditioning unit. The electrostatic field excitation unit is used to apply an alternating electric field to the electrostatic induction unit array, with an operating frequency of 10-100kHz and an excitation voltage of 5-30V; the signal conditioning unit is used to receive the capacitance change signal of the electrostatic induction unit, and the signal conditioning unit includes a signal amplification module, a digital filtering module and an analog-to-digital conversion module.

[0042] The controller further comprises: a data processing unit for analyzing the approach distance and approach speed of a human body; and an early warning control unit for outputting corresponding control instructions according to the danger level.

[0043] Example 2

[0044] like Figure 2 FIG. 1 is a flow chart of a sensing method for protecting a robot according to the present invention, wherein the sensing method comprises the following steps:

[0045] Step S1, applying an alternating electric field with a frequency of 10-100 kHz and a voltage of 5-30 V to the electrostatic induction unit array on the flexible base of the robot arm, and collecting the capacitance change signal of the electrostatic induction unit.

[0046] In practical applications, the electrostatic field excitation unit uses digital signal synthesis technology to generate the excitation signal. For example, a 50kHz sine wave signal with an amplitude of 15V can be selected to drive the sensing electrode through a Class D power amplifier. In order to improve the signal-to-noise ratio, orthogonal demodulation technology can be used to multiply the in-phase component and the orthogonal component of the received signal and the excitation signal respectively, and then low-pass filter to extract the capacitance change information. In addition, the system can also dynamically adjust the excitation frequency to adapt to different environmental interference.

[0047] Step S2, amplifying the capacitance change signal using a programmable gain amplifier, eliminating interference through digital filtering, performing digital sampling through an analog-to-digital converter, and calculating the human body approach distance through an equivalent capacitance model based on the processed capacitance change signal.

[0048] The signal processing adopts a multi-stage amplification and filtering scheme. First, a low-noise instrumentation amplifier is used for pre-amplification, and the gain can be adjusted programmably between 1-1000; then the power frequency interference and high-frequency noise are removed through a bandpass filter (the center frequency is the excitation frequency, and the bandwidth is 1kHz); finally, a 24-bit ΣΔ analog-to-digital converter is used for sampling, and the sampling rate is 1kHz. In actual tests, this scheme can detect a capacitance change of 0.1pF, which is equivalent to detecting a human body approaching within a distance of 50cm.

[0049] Step S3: Establish a risk level assessment model based on the human body's approaching distance, the angle between the human body's approaching direction and the moving direction of the robotic arm, and the human body's approaching speed, and send a warning signal based on the assessment result.

[0050] Based on the processed capacitance change signal Calculate the human body's approaching distance The equivalent capacitance model is: , where is the distance correction coefficient, and its value range is 0.8 - 1.2. is the vacuum permittivity, and A is the area of the induction unit array where the capacitance signal changes.

[0051] The risk level assessment model is: , where R is the risk level index, and its value range is 0 - 100. is the human body's approaching distance, with the unit of meter. is the angle between the human body's approaching direction and the moving direction of the robotic arm, with the unit of radian. is the human body's approaching speed, with the unit of m / s, and the weight coefficients , , are obtained as follows:

[0052] , , , where , are the maximum and minimum values of the effective detection distance, taking 0.2 meters and 0.05 meters respectively. is the approaching speed reference value, taking 10 m / s.

[0053] According to the magnitude of the risk level index R, the risk level is divided into:

[0054] Safe state: R ≤ 30, the robotic arm operates normally.

[0055] Warning state: 30 < R ≤ 70, reduce the operating speed of the robotic arm to 50%.

[0056] Dangerous state: R > 70, the robotic arm stops urgently.

[0057] The risk level assessment model comprehensively considers multiple key factors of human - machine interaction. For example, when a staff member approaches a moving robotic arm from the front (θ = 0°), moves to a distance of 15 cm from the robotic arm at a speed of 0.5 m / s, the calculated risk level index R is approximately 65 at this time, and the system will automatically reduce the speed of the robotic arm to 50%. If approaching to 10 cm or the speed suddenly increases to 1 m / s, the R value will exceed 70, triggering an emergency stop.

[0058] The correction factor is calculated by the multi-point collaborative sensing algorithm to correct the human body approach distance d. The expression of the correction factor is:

[0059]

[0060] Among them, ΔCi is the capacitance change value of the i-th adjacent sensing unit, ΔCmax is the maximum capacitance change value in the current detection cycle, k2 is the dynamic response coefficient, the value range is 0.1-0.3, dv / dt is the rate of change of the approach distance, the unit is meter / second; the corrected human body approach distance .

[0061] This embodiment can also introduce multi-sensor data fusion technology, combine the electrostatic induction signal with the data of other sensors (such as infrared ranging, ultrasonic, etc.), and use the Kalman filter algorithm for information fusion to further improve the ranging accuracy and reliability. Experiments show that this method can control the ranging error within ±5mm.

[0062] It should be noted that those skilled in the art should understand that various changes and equivalent substitutions may be made to the present invention without departing from the scope of the present invention. In addition, various modifications may be made to the present invention for specific situations or materials without departing from the scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed, but should include all embodiments falling within the scope of the claims of the present invention.

Claims

1. A sensing method for protecting a robot's sensing skin, wherein the sensing skin comprises a sensing skin wrapped around the outside of a robot arm, a controller electrically connected to the sensing skin, and a flexible base layer arranged on the robot arm; the sensing skin comprises, from inside to outside, a flexible base layer arranged on the robot arm, an array of electrostatic sensing units evenly distributed on the surface of the flexible base layer, and a partitioned protective layer for covering and protecting the sensing units; characterized in that: The induction method includes the following steps: Step S1: Apply an alternating electric field with a frequency of 10 - 100 kHz and a voltage of 5 - 30 V to the electrostatic induction unit array on the flexible base layer of the robotic arm, and collect the capacitance change signals of the electrostatic induction units; Step S2: Use a programmable gain amplifier to amplify the capacitance change signals, eliminate interference through digital filtering, perform digital sampling through an analog-to-digital converter, and calculate the human approaching distance based on the processed capacitance change signals through an equivalent capacitance model; The equivalent capacitance model for calculating the human body approach distance d based on the processed capacitance change signal ΔC is: Wherein, k1 is the distance correction coefficient, the value range is 0.8-1.2, ε0 is the vacuum dielectric constant, and A is the area of ​​the sensing unit array where the capacitance signal changes; Step S3: Establish a risk level assessment model according to the human approaching distance, the angle between the human approaching direction and the movement direction of the robotic arm, and the human approaching speed, and issue a warning signal based on the assessment result; The hazard level assessment model is: Among them, R is the danger level index, ranging from 0 to 100, d is the human approach distance, in meters, θ is the angle between the human approach direction and the robot arm movement direction, in radians, v is the human approach speed, in meters per second, and the weight coefficients w1, w2, and w3 are determined as follows: Where dmax and dmin are the maximum and minimum values ​​of the effective detection distance, which are 0.2 meters and 0.05 meters respectively, and v′ is the reference value of the approach speed, which is 10 meters per second.

2. The sensing method according to claim 1, characterized in that: According to the magnitude of the risk level index R, the risk level is divided into: Safe state: R ≤ 30, the robotic arm operates normally; Warning state: 30 < R ≤ 70, reduce the operating speed of the robotic arm to 50%; Dangerous state: R > 70, the robotic arm stops urgently.

3. The sensing method according to claim 2, characterized in that: Calculate a correction factor through a multi-point collaborative perception algorithm to correct the human approaching distance d, and the expression of the correction factor is: where ΔCi is the capacitance change value of the i-th adjacent induction unit, ΔCmax is the maximum capacitance change value within the current detection period, k2 is the dynamic response coefficient, and its value range is 0.1 - 0.3, dv / dt is the change rate of the approaching distance, with the unit of meter per second; The corrected human approaching distance d′ = dF.

4. The sensing method according to claim 1, characterized in that: The flexible base layer includes a flexible circuit board made of polyimide material, with a thickness of 50 - 200 microns; an interconnected wire network is provided on the surface of the flexible circuit board, with a wire width of 50 - 100 microns, for connecting the electrostatic induction units; A conductive shielding layer is provided at the bottom of the flexible circuit board for suppressing electromagnetic interference from the robotic arm body; The electrostatic induction unit array includes metal electrodes, and the metal electrodes are made of copper, aluminum or silver conductive materials, with a thickness of 1 - 5 microns, and the electrode size is 2 mm × 2 mm to 5 mm × 5 mm; The electrode spacing is 0.1 - 0.5 mm; The partitioned protective layer includes: a plurality of independent silicone blocks, with a thickness of 0.5 - 2 mm; There are gaps between adjacent silicone blocks, and the gap width is 0.5 - 1.5 mm; the gaps are filled with conductive fabric for electrostatic shielding.

5. The sensing method according to claim 4, characterized in that: The controller includes: an electrostatic field excitation unit and a signal conditioning unit. The electrostatic field excitation unit is used to apply an alternating electric field to the electrostatic induction unit array, with a working frequency of 10 - 100 kHz and an excitation voltage of 5 - 30 V; the controller also includes: a data processing unit for analyzing the human approaching distance and approaching speed; a warning control unit for outputting corresponding control instructions according to the risk level; The signal conditioning unit is used to receive the capacitance change signals of the electrostatic induction units, and the signal conditioning unit includes a signal amplification module, a digital filtering module and an analog-to-digital conversion module.

Images