Humanoid robot bionic skin heating temperature control method

Abstract

The invention discloses a humanoid robot bionic skin heating temperature control method, and belongs to the technical field of robot automatic control, and the method comprises the following steps: obtaining real-time temperature distribution data of a bionic skin surface, carrying out time domain differential processing to generate temperature change rate data, and combining the obtained real-time pressure distribution data to construct a contact surface heat conduction characteristic; and generating a basic heat energy supply strategy based on the steady-state components in the heat conduction characteristics of the contact surface, performing dynamic weighted fusion on the basic heat energy supply strategy and an instantaneous heat flow disturbance compensation amount calculated based on the instantaneous dynamic components in the heat conduction characteristics of the contact surface, generating a control heating instruction, and driving a heating unit in the bionic skin to execute a heating action. The real-time pressure and the temperature change rate are integrated to construct the heat conduction characteristic of the contact surface, and the characteristic is decomposed into a steady-state component and an instantaneous dynamic component for feed-forward compensation control, so that predictive compensation can be performed on external thermal disturbance, and rapid and stable control on the temperature of the bionic skin is realized.

Description

Technical Field

[0001] This invention relates to the field of robot automatic control technology, and in particular to a method for controlling the temperature of bionic skin heating in humanoid robots. Background Technology

[0002] As a crucial development direction for advanced automation systems, humanoid robots are vitally important for their ability to interact safely and naturally with the environment and humans. Bionic skin, a key component covering the robot's surface to simulate the sensory functions of human skin, integrates multiple sensors to acquire information such as touch, pressure, and temperature. Maintaining a constant temperature close to human body temperature on the surface of the bionic skin is a critical technology in its design and control, aiming to enhance the realism and approachability of humanoid robot interactions.

[0003] Currently, for temperature control of bionic skin, existing technologies typically employ a single feedback control scheme based on temperature sensors. This type of scheme measures the real-time surface temperature using temperature sensors distributed throughout the skin and compares it to a preset target temperature. Using this temperature deviation as input, classic feedback control algorithms such as proportional-integral-derivative (PID) controllers are employed to adjust the power output of the built-in heating unit, thereby gradually eliminating the passively measured temperature deviation.

[0004] However, the aforementioned existing technical solutions have obvious inherent defects. Because their control logic relies entirely on the generated temperature deviation, there is an inherent lag in response. When the bionic skin comes into contact with an object with high thermal conductivity, such as metal, heat is rapidly dissipated, but the control system must wait for the surface temperature to drop significantly before it can begin to respond, resulting in poor real-time temperature control. Furthermore, simple feedback control cannot distinguish the source and nature of disturbances. For rapid, severe thermal disturbances, its adjustment is often not fast enough, easily leading to large temperature fluctuations. Increasing the controller gain to improve response speed can easily cause system overshoot and oscillations, affecting the stability of temperature control. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a method for heating and controlling the temperature of bionic skin in humanoid robots. By integrating real-time pressure and temperature change rates to construct the heat conduction characteristics of the contact surface, and decomposing these characteristics into steady-state and instantaneous dynamic components for feedforward compensation control, this method can provide predictive compensation for external thermal disturbances, thereby achieving rapid and stable control of the bionic skin temperature.

[0006] The above objectives can be achieved through the following approach: A method for heating and controlling the temperature of bionic skin in a humanoid robot includes: acquiring real-time temperature distribution data and real-time pressure distribution data of the bionic skin surface; performing time-domain differential processing on the real-time temperature distribution data to generate temperature change rate data; integrating the real-time pressure distribution data and the temperature change rate data to construct a contact surface heat conduction characteristic that characterizes the dynamic properties of heat exchange at the skin-external interface; generating a basic heat energy supply strategy to maintain the skin's basic thermal balance based on the steady-state components representing slow and continuous heat exchange in the contact surface heat conduction characteristic; calculating an instantaneous heat flow disturbance compensation amount to quickly offset sudden heat changes based on the instantaneous dynamic components representing rapid and sudden heat exchange in the contact surface heat conduction characteristic; dynamically weighting and fusing the basic heat energy supply strategy and the instantaneous heat flow disturbance compensation amount to generate a regulating heating command; and driving the heating unit within the bionic skin to perform heating actions according to the regulating heating command.

[0007] Optionally, the step of acquiring real-time temperature distribution data and real-time pressure distribution data of the bionic skin surface, and performing time-domain differential processing on the real-time temperature distribution data to generate temperature change rate data includes: acquiring real-time temperature distribution data and real-time pressure distribution data of the bionic skin surface; constructing time series data from the real-time temperature distribution data; calculating the temperature difference between adjacent time points in the time series data to generate a temperature change gradient; and generating temperature change rate data based on the temperature change gradient and time interval.

[0008] Optionally, the step of integrating the real-time pressure distribution data and the temperature change rate data to construct a contact surface thermal conduction feature characterizing the dynamic properties of heat exchange at the skin-external contact interface includes: identifying the pressure change pattern of the contact area from the real-time pressure distribution data; weighting the temperature change rate data in conjunction with the pressure change pattern to calculate the heat conduction rate; and integrating the pressure change pattern and the heat conduction rate to generate the contact surface thermal conduction feature.

[0009] Optionally, the step of generating a basic thermal energy supply strategy for maintaining the skin's basic thermal balance based on the steady-state components representing slow and continuous heat exchange in the contact surface thermal conduction characteristics includes: performing low-pass filtering on the contact surface thermal conduction characteristics to separate the steady-state components representing slow and continuous heat exchange; calculating the average heat demand characterizing continuous heat exchange based on the steady-state components; and generating a basic thermal energy supply strategy for maintaining the skin's basic thermal balance based on the average heat demand.

[0010] Optionally, the step of calculating the instantaneous heat flow disturbance compensation amount for rapidly offsetting sudden heat changes based on the instantaneous dynamic components representing rapid and sudden heat exchange in the contact surface heat conduction characteristics includes: subtracting the steady-state components from the contact surface heat conduction characteristics to extract the instantaneous dynamic components representing rapid and sudden heat exchange; comparing the amplitude of the instantaneous dynamic components with a preset disturbance threshold to identify heat mutation events; and calculating the instantaneous heat flow disturbance compensation amount for rapidly offsetting sudden heat changes based on the quantized amplitude of the heat mutation events.

[0011] Optionally, the step of dynamically weighting and fusing the basic thermal energy supply strategy with the instantaneous heat flow disturbance compensation to generate a control heating command includes: generating a first dynamic weight for the basic thermal energy supply strategy based on the real-time pressure distribution data; generating a second dynamic weight for the instantaneous heat flow disturbance compensation based on the temperature change rate data; and using the first dynamic weight and the second dynamic weight to perform a weighted summation of the basic thermal energy supply strategy and the instantaneous heat flow disturbance compensation to generate a control heating command.

[0012] Optionally, driving the heating unit within the bionic skin to perform a heating action according to the heating control command includes: parsing heating parameters from the heating control command; controlling the power output of the heating unit to perform the heating action according to the heating parameters; monitoring the actual execution state of the heating unit, and making closed-loop adjustments to the power output based on the monitoring results.

[0013] Optionally, the method includes: acquiring ambient temperature data of the robot's environment; modifying the heat conduction characteristics of the contact surface based on the ambient temperature data to generate modified heat conduction characteristics of the contact surface; and updating the basic heat energy supply strategy and the instantaneous heat flow disturbance compensation amount based on the modified heat conduction characteristics of the contact surface.

[0014] Optionally, the method further includes: defining a target temperature range for the bionic skin; comparing the real-time temperature distribution data with the target temperature range to generate temperature deviation data; and optimizing the heating control command based on the temperature deviation data.

[0015] Based on the same inventive concept, this invention also provides a humanoid robot bionic skin heating and temperature control system. The system further includes: a data acquisition module for acquiring real-time temperature distribution data and real-time pressure distribution data of the bionic skin surface, and performing time-domain differential processing on the real-time temperature distribution data to generate temperature change rate data; a heat conduction analysis module for integrating the real-time pressure distribution data and the temperature change rate data to construct a contact surface heat conduction characteristic that characterizes the dynamic properties of heat exchange at the skin-external interface; a heating decision module for generating a basic heat energy supply strategy to maintain the skin's basic thermal balance based on the steady-state components representing slow and continuous heat exchange in the contact surface heat conduction characteristic; a heating compensation module for calculating an instantaneous heat flow disturbance compensation amount to quickly offset sudden heat changes based on the instantaneous dynamic components representing rapid and sudden heat exchange in the contact surface heat conduction characteristic; a heating execution module for dynamically weighting and fusing the basic heat energy supply strategy with the instantaneous heat flow disturbance compensation amount to generate a regulating heating command; and driving the heating unit within the bionic skin to perform heating actions according to the regulating heating command.

[0016] Compared with the prior art, the present invention has the following advantages: 1. This invention constructs a feedforward compensation control mechanism based on multimodal sensing information, which can predictively adjust heat exchange caused by external contact; by integrating pressure distribution data and temperature change rate data, the system can estimate the dynamic characteristics of heat exchange before thermal disturbances significantly affect the skin surface temperature, and generate compensatory heating commands in advance, thereby greatly shortening the system's response delay, suppressing temperature overshoot and fluctuations, and achieving rapid dynamic response; 2. This invention proposes a control strategy that decomposes the dynamic characteristics of heat exchange into steady-state components and instantaneous dynamic components, enabling differentiated handling of thermal disturbances of different natures. By providing basic heat energy supply for slow and continuous background heat exchange, while providing instantaneous heat flow disturbance compensation for rapid and sudden heat changes, and dynamically weighting and fusing the two, the control system can achieve an intelligent balance between maintaining long-term temperature stability and achieving rapid instantaneous response, thus improving the precision and adaptability of control. 3. This invention constructs a composite control architecture combining feedforward and feedback by introducing ambient temperature for model correction and introducing actual temperature deviation for feedback optimization, thereby improving the robustness and control accuracy of the system. The introduction of ambient temperature enables the system to actively adapt to changes in the external environment, while the feedback optimization of temperature deviation can effectively compensate for model inaccuracies and unmodeled disturbances, ensuring that the skin temperature can accurately converge and stabilize within the target range under various complex working conditions, thus guaranteeing the overall performance of the temperature control system.

[0017] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the humanoid robot bionic skin heating and temperature control method according to an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the structure of the humanoid robot bionic skin heating and temperature control system according to an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Reference Figure 1 One embodiment of the present invention proposes a method for heating and controlling the temperature of bionic skin of a humanoid robot. By integrating real-time pressure and temperature change rate to construct the heat conduction characteristics of the contact surface, and decomposing the characteristics into steady-state components and instantaneous dynamic components for feedforward compensation control, it is possible to predictively compensate for external thermal disturbances and achieve rapid and stable control of the temperature of the bionic skin.

[0023] The method described in this embodiment specifically includes: Real-time temperature distribution data and real-time pressure distribution data of the bionic skin surface are acquired, and the real-time temperature distribution data is processed by time-domain differentiation to generate temperature change rate data. By integrating the real-time pressure distribution data and the temperature change rate data, a contact surface heat conduction characteristic is constructed to characterize the dynamic properties of heat exchange at the skin-external interface. Based on the steady-state components representing slow and continuous heat exchange in the heat conduction characteristics of the contact surface, a basic heat energy supply strategy for maintaining the basic thermal balance of the skin is generated. Based on the instantaneous dynamic components representing rapid and sudden heat exchange in the heat conduction characteristics of the contact surface, the instantaneous heat flow disturbance compensation amount for rapidly offsetting sudden heat changes is calculated. The basic thermal energy supply strategy and the instantaneous heat flow disturbance compensation are dynamically weighted and fused to generate a control heating command; According to the heating control command, the heating unit inside the bionic skin is driven to perform a heating action.

[0024] This invention first acquires skin surface pressure and temperature data synchronously, and then performs time-domain differentiation on the temperature data to transform static physical quantities into dynamic indicators characterizing the intensity of heat exchange, namely, the rate of temperature change. Subsequently, it integrates pressure data representing the physical properties of the contact with temperature change data representing the velocity of heat flow to construct a contact surface heat conduction characteristic that comprehensively describes the thermal dynamics of the contact interface. This characteristic is decomposed into a steady-state component representing background, continuous heat exchange, and an instantaneous dynamic component representing sudden, transient heat exchange. Based on this decomposition, the system generates two heating strategies in parallel: a basic heat energy supply strategy to maintain overall thermal balance, and an instantaneous heat flow disturbance compensation quantity to quickly counteract external disturbances. Finally, these two strategies are intelligently fused through dynamic weighting to form the final control heating command, thereby driving the heating unit to perform precise heating actions.

[0025] This method improves the dynamic response speed, stability, and environmental adaptability of the bionic skin temperature control system. By introducing pressure and temperature change rate as feedforward signals, the system can proactively compensate for thermal disturbances before the actual temperature deviates, thereby shortening the response delay and suppressing temperature overshoot and fluctuations. Decomposing heat exchange into steady-state and transient states for separate control allows the system to smoothly maintain a base temperature while responding to sudden thermal stimuli, avoiding the limitations of a single control strategy in different scenarios. Ultimately, this intelligent fusion control approach makes the temperature performance of the bionic skin more stable and natural, improving the realism of thermal sensation during interaction with the external environment and enhancing the overall robustness of the system in complex and variable environments.

[0026] Optionally, the step of acquiring real-time temperature distribution data and real-time pressure distribution data of the bionic skin surface, and performing time-domain differential processing on the real-time temperature distribution data to generate temperature change rate data includes: Acquire real-time temperature distribution data and real-time pressure distribution data on the surface of the bionic skin; Time series data are constructed from the real-time temperature distribution data; Calculate the temperature difference between adjacent time points in the time series data to generate a temperature change gradient; Temperature change rate data are generated based on the temperature change gradient and time interval.

[0027] Specifically, firstly, real-time temperature and pressure distribution data covering the entire skin area are simultaneously collected and acquired using temperature and pressure sensor arrays distributed across the surface of the humanoid robot's bionic skin. For any sensing node on the bionic skin, the system continuously records its temperature reading at a fixed sampling period, thus constructing a time-series data set from the discrete real-time temperature distribution data. Subsequently, to quantify the temperature change trend over time, the system calculates the temperature difference between two adjacent time points in the time-series data; this difference is the temperature gradient. This process can be expressed by the following formula: , In the formula, This refers to the temperature change rate data calculated at a specific sensing node. This represents the temperature value of the corresponding node in the real-time temperature distribution data at the current sampling time. This is the temperature value recorded at the previous sampling time. This is a fixed time interval between two samplings. By repeating this calculation on all temperature sensing nodes on the bionic skin, a complete set of temperature change rate data characterizing the dynamic changes in temperature across the entire bionic skin surface can be obtained.

[0028] By performing time-domain differentiation on real-time temperature distribution data, this method can transform static temperature readings into dynamic temperature change rate data. This processing method enables the control system to not only respond to the absolute value of temperature, but also to sense the speed and direction of temperature changes. This improves the system's sensitivity and response speed to external thermal disturbances, allowing it to distinguish between rapid, instantaneous heat exchange and slow, continuous heat exchange. This provides crucial dynamic input information for subsequently constructing contact surface heat conduction characteristics that accurately reflect the dynamic nature of heat exchange, thereby achieving predictive compensation and control of heat changes.

[0029] Optionally, the step of integrating the real-time pressure distribution data and the temperature change rate data to construct the contact surface heat conduction characteristics characterizing the dynamic properties of heat exchange at the skin-external interface includes: Identify the pressure change pattern in the contact area from the real-time pressure distribution data; The temperature change rate data is weighted and processed in conjunction with the pressure change pattern to calculate the heat conduction rate. The pressure change pattern and the heat conduction rate are integrated to generate the thermal conduction characteristics of the contact surface.

[0030] Specifically, the system first analyzes real-time pressure distribution data acquired from the sensor array. A pressure threshold is set to distinguish between contact and non-contact areas. Then, within the identified contact areas, the specific pressure values, distribution patterns, and their time-varying trends are collectively defined as a pressure change pattern. This pressure change pattern physically reflects the tightness and nature of contact between the bionic skin and external objects. Next, to quantify the heat exchange caused by contact, the system combines the previously calculated temperature change rate data with this pressure change pattern. Specifically, the system uses a pre-calibrated or modeled pressure-dependent thermal coupling function to weight the temperature change rate data, thereby calculating the heat conduction rate. This calculation process can be expressed by the following formula: , In the formula, The calculated heat conduction rate is physically represented as the heat flux density per unit area. It is a pressure-related thermal coupling coefficient that characterizes the equivalent thermal conductivity of the contact interface under a specific pressure P. Its value is obtained through experimental calibration or based on contact thermal resistance theory modeling. P is the local pressure value extracted from the real-time pressure distribution data. This provides the temperature change rate data for the corresponding location. Finally, the system compares the pressure change pattern, i.e., the pressure value P, with the calculated heat transfer rate. These features are integrated to form a two-dimensional or multi-dimensional feature vector, which is the final generated contact surface heat conduction feature.

[0031] Among them, the pressure-related thermal coupling coefficient The core of this method is to quantify the equivalent thermal conductivity of the contact interface under different pressures. Those skilled in the art can determine its specific form through the following methods. This method establishes the coupling coefficient between pressure P and thermal conductivity through controlled experiments. The correspondence between them is established. Prepare a bionic skin sample, a standard test block with known heat capacity and thermal conductivity (e.g., an aluminum block), a precision heat flow sensor, a pressure sensor, and a temperature control device. Bring the bionic skin into contact with the standard test block, and apply a series of discrete, precise pressure values ​​using a force-applying device. (For example: 0.1, 0.5, 1.0, 2.0, 5.0 N / cm²). At each pressure... This creates a temperature difference between the skin and the object, while simultaneously using a heat flux sensor to measure the actual heat flux density at the interface. And record the rate of temperature change of the skin surface during this process. According to the formula Calculate each pressure The corresponding thermal coupling coefficient The obtained data points By performing function fitting, a continuous function is obtained. For example, a piecewise linear function can be used for fitting.

[0032] For example, experimental calibration has shown that for typical silicone bionic skin, its It can be approximated as a piecewise function as follows: , In practical applications, the controller can determine the current pressure P's range and substitute it into the corresponding formula to calculate... Alternatively, it can be obtained directly through a pre-stored lookup table.

[0033] By integrating pressure and temperature change information, this method establishes a causal understanding of the heat exchange process, enabling the system to distinguish heat exchange caused by different physical contacts. For example, it can differentiate between rapid heat loss due to close contact with a highly thermally conductive object and slow heat change caused by slight contact with air or a low-thermal-conductivity object. Through the calculated heat conduction rate, the system transforms qualitative temperature changes into quantitative heat flux, providing a physical basis for subsequent energy compensation. The resulting contact surface thermal conduction characteristics provide the control system with a description of the thermal dynamics of the contact interface, improving the targeting and accuracy of the heat control strategy.

[0034] Optionally, the strategy for generating a basic thermal energy supply to maintain the skin's basal thermal balance, based on the steady-state components representing slow and continuous heat exchange in the thermal conduction characteristics of the contact surface, includes: The thermal conduction characteristics of the contact surface are subjected to low-pass filtering to separate the steady-state components representing slow and continuous heat exchange; The average heat demand characterizing continuous heat exchange is calculated based on the aforementioned steady-state components. A basic thermal energy supply strategy for maintaining the skin's basic thermal balance is generated based on the average thermal demand.

[0035] Specifically, the system first receives the previously constructed contact surface thermal conduction characteristics, which characterize the dynamic properties of heat exchange at the skin-external interface. This characteristic is a time-series data set containing the heat conduction rate that varies over time. To extract the portion representing slow and continuous heat exchange, the system performs a digital low-pass filter on this time-series data of heat conduction rate. This filtering operation aims to smooth out high-frequency noise and brief bursts in the data, thereby separating the baseline trend of the signal, which is the steady-state component. A typical implementation of digital low-pass filtering can be expressed by the following equation: , In the formula, This represents the steady-state composition calculated at the current moment. This represents the heat conduction rate obtained from the thermal conduction characteristics of the contact surface at the current moment. This is the steady-state component value calculated at the previous moment. This is a filter coefficient between 0 and 1, and its value determines the smoothness of the filter. A smaller value means a stronger smoothing effect and can more effectively filter out rapid changes. This steady-state component Numerically, it directly reflects the average heat loss or gain that occurs continuously under the current contact state; therefore, it is directly used to characterize the average heat demand for continuous heat exchange. Finally, based on this average heat demand, the system generates a basic thermal energy supply strategy to maintain the skin's basal thermal balance. The core of this strategy is to provide an energy supply equal in magnitude but opposite in direction to the average heat demand, in order to achieve dynamic equilibrium. Therefore, the aforementioned basic thermal energy supply strategy... It can be directly set to equal the calculated average heat demand. .

[0036] Among them, the filter coefficients The value of determines the criteria for classifying the system as "steady state" and "transient state," and its selection should be related to the system's sampling period. The characteristic time constant of the signal to be separated from the desired signal Related. The equivalent time constant of this first-order low-pass filter. Approximately equal to Time constant The physical meaning is: the system considers the duration to be longer than The heat exchange process is in a steady state and is shorter than [a certain value]. This is considered transient. Those skilled in the art should select the appropriate timescale based on the typical timescale of the robot interaction scenario. to determine In humanoid robot interactions, sudden contact such as touching or tapping typically occurs within 1 second, while sustained contact such as holding or leaning usually lasts longer than 1-2 seconds. Therefore, a reasonable distinction time point can be set between 0.5 and 2 seconds. Assuming the system sampling frequency is 100Hz, the sampling period is 0.01 seconds. If heat exchange lasting longer than 1 second is considered a steady state, then a time constant can be set. seconds. Then the filter coefficients... To maintain stability even with shorter disturbances, you can set... Seconds, at this moment In typical robot interaction applications, The preferred range is between 0.005 and 0.05. Smaller... Values ​​(such as 0.005) produce a smoother steady-state component and are more effective at suppressing transient disturbances, but they are slower at following steady-state changes. Larger values... Values ​​(such as 0.05) follow steady-state changes more quickly, but allow some shorter disturbances to mix into the steady-state component.

[0037] By applying low-pass filtering to the thermal conduction characteristics of the contact surface, this method decomposes the dynamic characteristics of heat exchange into steady-state and transient components, providing a stable and continuous basic heating command for the biomimetic skin's temperature control system. This strategy aims to offset background heat loss caused by the environment and continuous contact, thereby establishing an energy balance benchmark for maintaining the overall base temperature of the skin. This approach avoids the control system overreacting to every minute thermal fluctuation, improving not only the smoothness and comfort of temperature control but also laying the foundation for rapid compensation for sudden thermal disturbances, making the entire control system more robust and efficient.

[0038] Optionally, the calculation of the instantaneous heat flow disturbance compensation amount for rapidly offsetting sudden heat changes, based on the instantaneous dynamic components representing rapid and sudden heat exchange in the heat conduction characteristics of the contact surface, includes: Subtract the steady-state component from the thermal conduction characteristics of the contact surface to extract the instantaneous dynamic component representing rapid and sudden heat exchange; The amplitude of the instantaneous dynamic component is compared with a preset disturbance threshold to identify thermal abrupt events; The instantaneous heat flow disturbance compensation amount is calculated based on the quantized amplitude of the heat mutation event to quickly offset the sudden heat change.

[0039] Specifically, the system first subtracts the separated steady-state components from the complete contact surface heat conduction characteristics to extract the instantaneous dynamic components representing rapid and sudden heat exchange. This process can be described by the following equation: , In the formula, The instantaneous dynamic component calculated at the current moment represents the rapidly fluctuating portion of the heat exchange process. This represents the total heat conduction rate obtained from the thermal conduction characteristics of the contact surface at the current moment. This is the steady-state component obtained after low-pass filtering. This subtraction operation is physically equivalent to decomposing the total heat exchange into a slow background component and a fast disturbance component. Next, to determine whether a heat change constitutes a sudden event requiring immediate response, the system compares the amplitude, i.e., the absolute value, of the instantaneous dynamic component with a pre-set disturbance threshold. Only when the amplitude of the instantaneous dynamic component exceeds this threshold is the system identified as a significant heat abrupt change event. Finally, once a heat abrupt change event is identified, the system will determine the quantized amplitude of the event, i.e., the instantaneous dynamic component... The instantaneous heat flux disturbance compensation amount is calculated based on its own value to rapidly offset sudden changes in heat. In specific implementation, this instantaneous heat flux disturbance compensation amount... It can be directly set to be equal to the identified instantaneous dynamic component. Its goal is to generate a compensating heat flow that is equal in magnitude but opposite in direction to the sudden change in heat.

[0040] By separating and quantifying instantaneous dynamic components from the total heat exchange, a feedforward disturbance compensation mechanism is achieved, enabling the temperature control system to respond extremely quickly and accurately to sudden heat exchange events, such as sudden contact with cold metal or heat sources. The system no longer needs to wait for a significant change in skin surface temperature before passively adjusting; instead, it actively compensates the moment it senses a sudden change in heat flow. This proactive disturbance cancellation strategy shortens the system's response time, suppresses temperature overshoot and fluctuations caused by external abrupt changes, ensures the temperature stability of the bionic skin during interaction with variable environments, and achieves a high degree of simulation of the physiological response of human skin to rapid thermal stimulation.

[0041] Optionally, the step of dynamically weighting and fusing the basic thermal energy supply strategy with the instantaneous heat flow disturbance compensation to generate a control heating command includes: A first dynamic weight for the basic thermal energy supply strategy is generated based on the real-time pressure distribution data; A second dynamic weight is generated based on the temperature change rate data for the compensation amount of the instantaneous heat flow disturbance; The basic thermal energy supply strategy and the instantaneous heat flow disturbance compensation are weighted and summed using the first dynamic weight and the second dynamic weight to generate a control heating command.

[0042] Specifically, after obtaining the basic thermal energy supply strategy for maintaining the skin's basic thermal balance and the instantaneous heat flow disturbance compensation for offsetting sudden heat changes, the system combines the two using a dynamic weighted fusion method to generate the final regulating heating command, dynamically adjusting the contribution ratio of the two components based on real-time sensing data. First, the system generates a first dynamic weight for the basic thermal energy supply strategy based on real-time pressure distribution data obtained from the pressure sensor array. This weight aims to reflect the stability of the contact state, typically assigning a higher weight value under no-contact or stable contact conditions to emphasize the importance of maintaining basic thermal balance. Second, the system generates a second dynamic weight for the instantaneous heat flow disturbance compensation based on previously calculated temperature change rate data. This weight aims to reflect the intensity of heat exchange; its weight value increases as the absolute value of the temperature change rate significantly increases, prioritizing responses to rapid heat surge events. Finally, the system generates the final regulating heating command by weighted summing of the two strategies and their corresponding dynamic weights. This fusion process can be represented by the following formula: , In the formula, The final generated control heating command is expressed in physical units as heat flux density. This is the basic thermal energy supply strategy. This is the compensation amount for the instantaneous heat flow disturbance. The first dynamic weight is a dimensionless coefficient determined by real-time pressure distribution data, used to adjust the strength of the basic supply strategy. , In the formula, This is the contact pressure value. This is the pressure threshold; below this value, it is considered non-contact or ineffective contact. Based on the sensitivity of the bionic skin, it can be preferably set to... , The pressure gain coefficient determines how quickly the weight transitions from 0 to 1. It is an exponential function. To achieve a more defined switching effect, it can be preferably set to 20. When P < 0.2 N / cm², the first dynamic weight is very close to 0. The item is almost ineffective because, in the absence of stable contact, the primary heat exchange comes from convection with the environment. The second dynamic weight is a dimensionless coefficient determined by the temperature change rate data, used to adjust the strength of the instantaneous compensation strategy. , In the formula, This represents the absolute value of the current rate of temperature change, in °C / s. The heat threshold is used to determine whether a sudden heat event has occurred, with a preferred value of 2.0°C / s. The temperature gain coefficient, preferably 3.0, provides a relatively smooth but responsive transition.

[0043] By dynamically weighting and fusing the basic heat energy supply strategy with instantaneous heat flow disturbance compensation, the temperature control system achieves a dynamic balance between maintaining long-term temperature stability and achieving rapid instantaneous response. When there is no interaction with the external environment or in a stable contact state, the system prioritizes executing a stable basic heat energy supply strategy to ensure smooth temperature distribution and optimized energy consumption. However, in the event of a rapid heat exchange event, the system can immediately increase the weight of the instantaneous compensation, mobilizing more energy for rapid and precise counter-heating or stopping heating, thereby suppressing the disturbance before it causes temperature deviations. This fusion mechanism ensures that the final heating control is neither too sluggish nor too sensitive, achieving precise matching and efficient management of heat demands under different interaction scenarios.

[0044] Optionally, driving the heating unit within the bionic skin to perform a heating action according to the controlled heating command includes: The heating parameters are parsed from the heating control command; The power output of the heating unit is controlled according to the heating parameters to perform the heating action; The actual operating status of the heating unit is monitored, and the power output is adjusted in a closed loop based on the monitoring results.

[0045] Specifically, first, the system receives a heating control command generated by the upper-layer fusion module. This command is essentially a numerical value specifying a target heat flux density. The system then parses the heating parameters for each individual heating unit from this command. This parsing process requires converting the heat flux density command into a specific power command, calculated as follows: , In the formula, These are the analyzed heating parameters, representing the target power output, in watts. The target heat flux density is obtained from the heating control command. This refers to the effective area of ​​the bionic skin covered by a single heating unit. The system then outputs power based on this target. The system controls the driving circuit of the heating unit to perform the heating action. A common implementation method is to adjust the average power applied to the heating unit using pulse width modulation (PWM). To ensure accurate execution of commands, the system monitors the actual operating status of the heating unit in real time. This is typically achieved by measuring the actual current flowing through the heating unit and the voltage across its terminals, thereby calculating the actual power output. Finally, based on the monitored actual power output, the system performs closed-loop adjustment of the control signal for the driving circuit. Specifically, a local controller continuously compares the difference between the target power output and the actual power output, and dynamically fine-tunes the duty cycle of the PWM based on this difference to ensure that the actual applied heating power accurately and stably tracks the target value required by the upper-level command.

[0046] By translating heating commands into physical power output and introducing a closed-loop adjustment mechanism, this method ensures the final execution accuracy of the entire temperature control strategy, constructing a reliable execution layer that faithfully reproduces the precise thermal energy requirements calculated by the upper-level algorithm into the physical world. This closed-loop power control method overcomes the interference of uncertainties such as power supply voltage fluctuations and temperature-dependent changes in heating unit resistance on the actual heating effect, ensuring that the energy output of each heating action is controllable and predictable. This improves the stability and reliability of the entire temperature control system, ensuring that high-level thermal balance and disturbance compensation strategies can be implemented without compromise, thereby ultimately guaranteeing the overall performance of bionic skin temperature control.

[0047] Optionally, the method includes: Obtain ambient temperature data of the robot's environment; The thermal conductivity characteristics of the contact surface are corrected by combining the ambient temperature data to generate the corrected thermal conductivity characteristics of the contact surface. Based on the corrected contact surface heat conduction characteristics, the basic heat energy supply strategy and the instantaneous heat flow disturbance compensation amount are updated.

[0048] Specifically, the system first acquires real-time ambient temperature data of the robot's external environment using dedicated temperature sensors deployed in the non-contact areas of the humanoid robot. Next, the system corrects the previously constructed heat conduction characteristics of the contact surface. This correction process aims to take into account the background heat exchange between the bionic skin and the surrounding environment through convection and radiation. Specifically, the system needs to calculate an environmental heat exchange component, which is calculated as follows: , In the formula, This is the environmental heat exchange component, representing the heat flux density between the skin and the environment per unit area. It is a comprehensive convective heat transfer coefficient, the value of which is pre-calibrated or dynamically estimated based on conditions such as airflow velocity on the robot surface. This refers to the local temperature of the bionic skin surface obtained from real-time temperature distribution data. This refers to the newly acquired ambient temperature data. The system then superimposes this ambient heat exchange component onto the existing contact surface heat conduction characteristics to generate a corrected contact surface heat conduction characteristic. This corrected characteristic more comprehensively reflects the total heat exchange on the skin surface. Finally, the system uses this corrected contact surface heat conduction characteristic as new input to re-execute the subsequent calculation process: low-pass filtering is performed again to update the basic heat energy supply strategy for maintaining the skin's basic thermal balance, and high-pass separation is performed again to update the instantaneous heat flow disturbance compensation amount used to quickly offset sudden changes in heat.

[0049] The value of h is closely related to the airflow velocity on the robot's surface. Different h values ​​can be set for several typical working states through offline calibration or by consulting relevant manuals, and the robot's own state can be switched accordingly. In a static state, such as when the robot is standing still, the air is in natural convection, and h can be set to 5 W / (m²·K); in a low-speed motion state, such as when the arm moves slowly, the air is in low-speed laminar flow, and h can be set to 15 W / (m²·K); in a high-speed motion state, such as when the arm swings rapidly, the air is in turbulent flow, and h can be set to 30 W / (m²·K). The robot's main control system can determine the current motion state of the arm based on the joint velocities output by its motion planning module and select the corresponding h value.

[0050] By introducing ambient temperature as a global variable, the environmental adaptability and control precision of the temperature control system are improved, enabling the system to clearly distinguish between heat exchange caused by object contact and background heat exchange caused by changes in ambient temperature. This makes the calculation of the basic thermal energy supply strategy more accurate, as it can now actively compensate for continuous heat loss or gain caused by environmental changes, such as when a robot moves from indoors to outdoors. This environmental compensation mechanism avoids slow overall temperature drift in response to environmental changes, not only enhancing the robustness of the bionic skin working in different thermal environments but also further optimizing the overall energy efficiency of the system by providing a baseline energy requirement.

[0051] Optionally, the method further includes: Define the target temperature range for bionic skin; The real-time temperature distribution data is compared with the target skin temperature range to generate temperature deviation data; The heating control command is optimized based on the temperature deviation data.

[0052] Specifically, the system first predefines the ideal working temperature of the bionic skin, i.e., the target temperature range of the skin, which typically includes a target temperature setpoint. Next, the system compares the real-time temperature distribution data collected on the surface of the bionic skin with this target temperature setpoint point by point to calculate and generate temperature deviation data. This temperature deviation data reflects the difference between the current actual temperature and the desired temperature. This calculation process can be expressed by the following formula: , In the formula, This refers to the generated temperature deviation data. Set a predefined target skin temperature value. The current temperature value at a specific location is obtained from real-time temperature distribution data. This temperature deviation data is then fed into a feedback controller, such as a proportional-integral controller, to generate a feedback correction. Finally, the system superimposes this feedback correction onto the previously generated heating command from the feedforward path, thereby optimizing the command and generating the final heating command. The optimization process can be represented as: , In the formula, This is the final heating control command after feedback optimization. This is the initial control heating command generated solely based on the feedforward model. This represents the temperature deviation data at the current moment. This is the proportional gain coefficient. These are the integral gain coefficients. Both coefficients are preset controller parameters used to adjust the strength and response characteristics of the feedback correction. This represents the accumulation of historical temperature deviations and is used to eliminate steady-state errors in the system.

[0053] By introducing a feedback loop based on actual temperature, the robustness and final control accuracy of the entire temperature control system are improved. Feedforward control can quickly respond to predictable disturbances, while feedback optimization handles residual temperature deviations caused by inaccurate models, unmodeled disturbances, and various uncertainties. This feedback loop ensures that the temperature of the bionic skin accurately converges and stabilizes within the set target temperature range during long-term operation, eliminating steady-state errors. This composite control architecture combining feedforward and feedback gives the system both rapid dynamic response and steady-state maintenance capabilities, thus achieving comprehensive and reliable temperature control of the bionic skin.

[0054] Based on the same inventive concept, such as Figure 2 As shown, the present invention also provides a humanoid robot bionic skin heating and temperature control system, the system further comprising: The data acquisition module is used to acquire real-time temperature distribution data and real-time pressure distribution data of the bionic skin surface, and to perform time-domain differential processing on the real-time temperature distribution data to generate temperature change rate data. The heat conduction analysis module is used to integrate the real-time pressure distribution data and the temperature change rate data to construct the contact surface heat conduction characteristics that characterize the dynamic properties of heat exchange at the interface between the skin and the outside world. The heating decision module is used to generate a basic heat energy supply strategy for maintaining the basic thermal balance of the skin based on the steady-state components representing slow and continuous heat exchange in the heat conduction characteristics of the contact surface. The heating compensation module is used to calculate the instantaneous heat flow disturbance compensation amount to quickly offset sudden heat changes based on the instantaneous dynamic components representing rapid and sudden heat exchange in the heat conduction characteristics of the contact surface. The heating execution module is used to dynamically weight and fuse the basic heat energy supply strategy with the instantaneous heat flow disturbance compensation amount to generate a heating control command; and to drive the heating unit in the bionic skin to perform heating actions according to the heating control command.

[0055] To verify the feasibility of this invention in practice, it was applied to a robot. This robot operates in a simulated dynamic service environment such as a coffee shop or a high-end care center, requiring frequent physical contact with objects and people at varying temperatures. The temperature stability and thermal realism of its bionic skin are crucial to the user experience. Traditional temperature control methods rely solely on feedback adjustments based on temperature deviations, resulting in slow responses. This causes the robot's skin temperature to drop sharply when touching a cold cup, or to feel unnatural when shaking hands with a person.

[0056] In this embodiment, the bionic skin of the robot's hand integrates a high-density temperature sensor array and a pressure sensor array. The system synchronously acquires real-time temperature and pressure distribution data of the skin surface at a frequency of 100Hz, and immediately performs time-domain differentiation on the temperature data to generate accurate temperature change rate data. Based on this data, the system constructs the heat conduction characteristics of the contact surface and decomposes them into steady-state and instantaneous dynamic components, generating a basic heat energy supply strategy and an instantaneous heat flow disturbance compensation amount, respectively. Finally, a heating control command is generated through dynamic weighted fusion to drive the micro-heating unit within the skin to perform heating actions.

[0057] To verify the beneficial effects of this invention, three key scenarios were designed for testing, and the results were compared with those obtained using only traditional PID feedback control. The test scenarios included: the robot's hand grasping a glass of ice water with a surface temperature of 5°C; the robot shaking hands with a test subject with a body temperature of 36.5°C; and the robot moving from an air-conditioned room at 22°C to an outdoor balcony at 30°C. The robot's target skin temperature range was preset to 34°C to 36°C, with a target value of 35°C.

[0058] In tests involving grasping a glass of ice water, the pressure sensor immediately detected contact the moment the robot's finger touched the glass wall, while the temperature sensor simultaneously detected a sharp drop in temperature. The method of this invention utilizes calculated large negative temperature change rates and pressure data to construct the thermal conduction characteristics of the contact surface, incorporating significant heat loss. Based on this, the system immediately generates a powerful instantaneous heat flow disturbance compensation quantity and assigns it a very high weight, driving the heating unit to perform high-power pre-compensation heating under feedforward control. Results show that the skin temperature using this invention only dropped to a minimum of 32.1°C and rapidly recovered to 35°C within 1.8 seconds. In contrast, the traditional method only began heating after the temperature had significantly deviated from the target, causing the skin temperature to drop to 24.5°C and take more than 5 seconds to slowly recover.

[0059] In handshake tests, the method of this invention also sensed contact with a warm, highly thermally conductive object through pressure and temperature change rates. The system determined this was a heat input rather than loss scenario, with the instantaneous dynamic component showing a positive value. The system quickly set the instantaneous heat flow disturbance compensation to a negative value, reducing heating and lowering the weight of the basic heat supply strategy. This allows the robotic hand to quickly match human body temperature, avoiding the "feverish" feeling caused by continuous heating. Results showed that the hand temperature of this invention stabilized at around 36.2°C after contact, providing a natural feel. In contrast, traditional methods, due to the integral effect, continued heating in the initial contact phase, causing the skin temperature to briefly rise to 38°C, resulting in an unnatural heat sensation for the tester.

[0060] In the environment switching test, when the robot moved from an indoor temperature of 22°C to an outdoor temperature of 30°C, the system acquired ambient temperature data through an ambient temperature sensor and used this data to correct the heat conduction characteristics of the contact surface. Based on the corrected characteristics, the system automatically adjusted the base heat energy supply strategy used to maintain basic thermal balance. This allowed the robot to smoothly reduce the background heating power after entering a warmer environment, avoiding temperature overshoot.

[0061] It should be noted that the electrical connections between the various units described above do not necessarily represent direct or indirect connections. Any indirect connection method can be applied to the embodiments of the present invention as long as it achieves the purpose of the present invention. The above descriptions are merely exemplary embodiments of the present invention and should not be construed as limiting the scope of the present invention.

[0062] All equivalent changes and modifications made in accordance with the teachings of this invention are still within the scope of this invention. Those skilled in the art will readily conceive of other embodiments of this invention upon considering the specification and the disclosure of practical truth. This application is intended to cover any variations, uses, or adaptations of this invention that follow the general principles of this invention and include common knowledge or conventional techniques in the art not described herein.

Claims

1. A method for heating temperature control of a human type robot bionic skin, characterized in that, The method comprises: obtaining real-time temperature distribution data and real-time pressure distribution data of the bionic skin surface, and performing time-domain differential processing on the real-time temperature distribution data to generate temperature change rate data; integrating the real-time pressure distribution data and the temperature change rate data to construct a contact surface heat conduction feature representing the dynamic characteristics of heat exchange at the contact interface between the skin and the external environment; based on the steady-state component in the contact surface heat conduction feature representing slow and continuous heat exchange, generating a basic heat energy supply strategy for maintaining the basic heat balance of the skin; based on the transient dynamic component in the contact surface heat conduction feature representing rapid and sudden heat exchange, calculating a transient heat flow disturbance compensation amount for quickly offsetting sudden heat changes; dynamically weighting and fusing the basic heat energy supply strategy and the transient heat flow disturbance compensation amount to generate a control heating instruction; driving a heating unit in the bionic skin to perform a heating action according to the control heating instruction.

2. The human robot bionic skin heating temperature control method according to claim 1, wherein, The method comprises: obtaining real-time temperature distribution data and real-time pressure distribution data of the bionic skin surface; constructing time series data from the real-time temperature distribution data; calculating the temperature difference between adjacent time points in the time series data to generate a temperature change gradient; generating temperature change rate data according to the temperature change gradient and the time interval.

3. The human robot bionic skin heating temperature control method according to claim 2, characterized in that, The method comprises: identifying the pressure change pattern of the contact area from the real-time pressure distribution data; combining the pressure change pattern to calculate the heat conduction rate by weighting the temperature change rate data; integrating the pressure change pattern and the heat conduction rate to generate the contact surface heat conduction feature.

4. The human robot bionic skin heating temperature control method according to claim 3, characterized in that, The method comprises: performing low-pass filtering on the contact surface heat conduction feature to separate the steady-state component representing slow and continuous heat exchange; calculating the average heat demand representing continuous heat exchange according to the steady-state component; generating a basic heat energy supply strategy for maintaining the basic heat balance of the skin according to the average heat demand.

5. The human robot bionic skin heating temperature control method according to claim 4, characterized in that, The method comprises: subtracting the steady-state component from the contact surface heat conduction feature to extract the transient dynamic component representing rapid and sudden heat exchange; comparing the amplitude of the transient dynamic component with a preset disturbance threshold to identify a heat mutation event; calculating a transient heat flow disturbance compensation amount for quickly offsetting sudden heat changes according to the quantitative amplitude of the heat mutation event.

6. The human robot bionic skin heating temperature control method according to claim 5, characterized in that, The method comprises: generating a first dynamic weight for the base thermal energy supply strategy according to the real-time pressure distribution data; generating a second dynamic weight for the transient thermal flow disturbance compensation amount according to the temperature rate of change data; performing a weighted sum of the base thermal energy supply strategy and the transient thermal flow disturbance compensation amount using the first dynamic weight and the second dynamic weight to generate a regulation heating instruction.

7. The human robot bionic skin heating temperature control method according to claim 6, characterized in that, The driving of the heating unit in the bionic skin to perform a heating action according to the regulation heating instruction comprises: parsing a heating parameter from the regulation heating instruction; controlling a power output of the heating unit according to the heating parameter to perform a heating action; monitoring an actual execution state of the heating unit and performing a closed-loop adjustment on the power output based on the monitoring result.

8. The human robot bionic skin heating temperature control method according to claim 1, wherein, The method comprises: obtaining environmental temperature data of an environment in which a robot is located; correcting the contact surface heat conduction feature in combination with the environmental temperature data to generate a corrected contact surface heat conduction feature; updating the base thermal energy supply strategy and the transient thermal flow disturbance compensation amount based on the corrected contact surface heat conduction feature.

9. The human robot bionic skin heating temperature control method according to claim 1, wherein, The method further comprises: defining a skin target temperature range of the bionic skin; comparing the real-time temperature distribution data with the skin target temperature range to generate temperature deviation data; performing feedback optimization on the regulation heating instruction based on the temperature deviation data.

10. A humanoid robot bionic skin heating temperature control system, characterized in that, The system further comprises: a data acquisition module for obtaining real-time temperature distribution data and real-time pressure distribution data of a surface of the bionic skin, and performing time-domain differential processing on the real-time temperature distribution data to generate temperature rate of change data; a heat conduction analysis module for integrating the real-time pressure distribution data and the temperature rate of change data to construct a contact surface heat conduction feature representing dynamic characteristics of heat exchange at a contact interface between the skin and the outside world; a heat supply decision module for generating a base thermal energy supply strategy for maintaining a basic thermal balance of the skin based on a steady-state component representing slow and continuous heat exchange in the contact surface heat conduction feature; a heat supply compensation module for calculating a transient thermal flow disturbance compensation amount for quickly offsetting sudden changes in heat based on a transient dynamic component representing fast and sudden heat exchange in the contact surface heat conduction feature; a heat supply execution module for dynamically weighting and fusing the base thermal energy supply strategy and the transient thermal flow disturbance compensation amount to generate a regulation heating instruction, and driving a heating unit in the bionic skin to perform a heating action according to the regulation heating instruction.

Images