Moxibustion robot and temperature control method

Abstract

The invention discloses a moxibustion robot and a temperature control method. The device comprises a machine body and a mechanical arm, a fan, a filter unit and a smoke treatment box are arranged in the machine body, and a phototherapy module and a moxibustion module can be hung at the tail end of the mechanical arm; the tail end moxibustion module comprises a moxibustion shell, a primary filter element arranged in the moxibustion shell, a moxa stick arranged in the center of the primary filter element, a moxa stick combustion cavity located at the bottom of the primary filter element and matched with the primary filter element, an exhaust fan arranged in a primary exhaust port and an anti-scalding baffle, and a smoke conveying pipe connector can be connected to an air inlet of a fan. The device has the advantages that the device can be quickly disassembled and assembled in a fire-free state, flexible switching between manual hand holding and robot automatic moxibustion applying is achieved, and quick and safe butt joint is guaranteed in a buckle or slot mode; a dual-mode smoke removal structure is adopted, so that portable smoke removal and oxygenation can be completed by using an exhaust fan and a primary filter screen which are arranged at the tail end; in a large smoke amount occasion, more powerful two-stage smoke discharge and smell removal are realized.

Description

Technical Field

[0001] The present invention relates to a moxibustion robot and a temperature control method based on multimodal sensor fusion, nonlinear prediction temperature control algorithm and detachable terminal modular expansion, belonging to the technical field of medical robots and traditional Chinese medicine moxibustion therapy. Background Art

[0002] Traditional Chinese medicine moxibustion has been valued in rehabilitation and physical therapy for its wide clinical application. However, traditional manual moxibustion is highly dependent on operating experience, and the temperature and distance of moxa sticks during burning are difficult to maintain constant. The safety and efficacy of treatment are also easily affected by human factors. In recent years, with the continuous development of robotics technology, some research institutions and companies have begun to combine moxibustion with robotic arms to form a variety of "moxibustion robots" or automated equipment to reduce the manual burden and improve the consistency of moxibustion. However, these products or prototypes still face the following shortcomings in actual use:

[0003] (1) Insufficient temperature control accuracy: Traditional thresholds or simple PID adjustments are difficult to adapt to unstable combustion environments, and are prone to temperature overshoot or underfire, that is, temperature control lags or overshoots. Traditional thresholds and PID algorithms are difficult to cope with fluctuations in moxa stick combustion, and are prone to burns or insufficient temperature. (2) Single smoke exhaust method: Only using a small fan or a high-power smoke pipe makes it difficult to take into account both "portability" and "high smoke volume" scenarios. (3) Integrated terminal structure design and lack of detachable terminals: Most robots need to be turned off or the system stopped before parts can be replaced, which affects the continuity and flexibility of moxibustion. (4) Trajectory and temperature control are disconnected: They can often only perform fixed round trips or rotations, and lack the ability to adjust the distance or trajectory radius in real time according to temperature changes. It is also impossible to dynamically adjust the distance between the end of the robotic arm and the skin according to the real-time temperature, which poses a risk of burns. (5) Limited safety monitoring capabilities, lack of comprehensive processing of multimodal data such as body movement and smoke concentration, and difficulty in timely linkage and abnormal avoidance. In response to the above-mentioned technical problems, some patents have been published that propose solutions from different angles, but there are still limitations in terms of temperature control models, smoke removal methods or terminal structures. For example: the utility model patent with announcement number: CN219763951U, which specifically discloses that secondary combustion is achieved through upper and lower shell heating / combustion-supporting parts to reduce the amount of smoke. This structural form reduces the amount of smoke to a certain extent, but its effect is still poor and the temperature cannot be predicted; another example: the invention patent application with publication number: CN117085498A, which specifically discloses the use of high-temperature catalytic secondary combustion to purify flue gas, also only achieves a certain flue gas purification effect and does not have temperature predictability. It can be seen that in the prior art, various moxibustion robots are still difficult to achieve a balance in "temperature control", "smoke exhaust efficiency", "human-machine collaboration" and other aspects, and urgently need better technical solutions to improve. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an intelligent moxibustion robot and its control method based on multi-modal sensor fusion, non-linear predictive temperature control algorithm and detachable end-module expansion, which can perform predictive temperature control using multi-modal sensing data (body surface temperature, distance, smoke concentration, human posture, etc.), and combine "dual-mode smoke removal" and "detachable end module" to achieve comprehensive improvement of high-precision temperature control, flexible smoke removal and human-machine collaboration capabilities.

[0005] To solve the above technical problem, the intelligent moxibustion robot based on multi-modal sensor fusion, non-linear predictive temperature control algorithm and detachable end-module expansion of the present invention includes a body and a robotic arm. A fan, a filtering unit disposed at the air outlet of the fan, and a flue gas treatment box connected to the filtering unit through a section of smoke pipe are provided inside the body. The end of the robotic arm can be mounted with a light therapy module and a moxibustion module. The end moxibustion module includes a moxibustion outer shell having a smoke delivery pipe interface and a primary exhaust port, a primary filter disposed inside the moxibustion outer shell, a moxa stick disposed at the center of the primary filter, a moxa stick combustion chamber located at the bottom of the primary filter and cooperatively installed with the primary filter, an exhaust fan disposed inside the primary exhaust port and cooperating with the primary filter, and a heat insulation baffle disposed at the bottom end of the moxibustion outer shell. The smoke delivery pipe interface can be connected to the air inlet of the fan through a smoke delivery pipe to achieve deep suction and filtration of smoke.

[0006] A card slot is provided at the end of the robotic arm, and a buckle cooperating with the card slot is provided on the moxibustion outer shell, so that the end moxibustion module can be snap-connected to the card slot of the robotic arm through the buckle.

[0007] The moxibustion outer shell is made of heat-insulating material.

[0008] The primary filter is a multi-layer filter screen, and the exhaust fan is a DC brushless fan.

[0009] A smoke sensor for monitoring smoke concentration information is provided inside the body. The control signals of the smoke sensor and the fan are both input to a control unit located inside the body. When the upper limit value of the smoke volume or odor concentration is detected by the smoke sensor, the fan can be triggered to operate to enhance the smoke exhaust efficiency.

[0010] The smoke delivery pipe is made of flame-retardant material.

[0011] Universal wheels are provided at the bottom of the body.

[0012] A temperature control method based on the above moxibustion robot includes the following steps: (A) Establish a non-linear temperature differential equation

[0013] Adopt the following model

[0014]

[0015] Wherein:

[0016] α represents the heat intensity coefficient of the flame head, which can fluctuate within a certain range according to the type of moxa stick and the combustion state;

[0017] ·1 / h(t) p (p≥2) is used to simulate the heat transfer that decays rapidly with the increase of distance;

[0018] ·β(T - T env ) complies with Newton's law of cooling and is suitable for the physical law of heat dissipation due to the temperature difference between the human body surface and the outside world;

[0019] (B) Iterative regulation is carried out through discretization

[0020] In actual control, the system executes in a loop with a fixed step size Δt (for example, 5 - 20 ms);

[0021] o The finite difference method is adopted, as shown in Equation (2):

[0022]

[0023] Where h n represents the distance between the end of the robotic arm and the body surface at time t n . By iterating step by step, the temperature increment at the "next moment" can be predicted within a very short time.

[0024] If Δt increases, medium / high - order methods are used to solve the differential equation.

[0025] After updating the distance / temperature for each discrete step, the system immediately passes the new expected distance Δh parameter to the trajectory generation module, and actually moves the position of the end of the robotic arm through inverse kinematics and joint interpolation.

[0026] The advantages of the present invention are:

[0027] First, the present invention provides a moxibustion module with integrated exhaust fans, primary filters and other components and can be quickly disassembled without interrupting the fire, so that it can be quickly disassembled and assembled without interrupting the fire, and flexible switching between manual hand-held and robot automatic moxibustion can be achieved, especially by means of snaps or slots to ensure fast and safe docking; secondly, in terms of temperature control, a nonlinear differential equation model based on the combustion characteristics of moxa sticks and the heat dissipation law of the body surface is constructed, and the early prediction of future temperature changes is achieved through numerical integration, and the sensor data such as temperature, distance, and smoke are integrated into the control cycle, so that the terminal is actively raised or retreated before the fire head temperature is about to exceed the standard, thereby avoiding overshoot; thirdly, a dual-mode smoke removal structure is adopted: in general diagnosis and treatment or home scenes, the exhaust fan and primary filter provided by the terminal can be used to complete portable smoke removal and oxygenation; in situations with large smoke volume, a high-power filter device arranged in the body is connected through a flexible pipe to achieve more powerful secondary smoke exhaust and deodorization. These two smoke removal modes can be switched manually or automatically according to sensor data, taking into account the requirements of portability and high-efficiency filtration;

[0028] At the same time, the present invention supports a variety of traditional moxibustion techniques, such as "sparrow pecking moxibustion" and "spiral moxibustion", and embeds the correction amount (such as distance increment and radius adjustment) output by the nonlinear temperature control algorithm into the trajectory planning in real time, and realizes smooth and temperature-adaptive three-dimensional moxibustion action through inverse kinematics interpolation; in addition, the system is equipped with a multimodal sensor fusion and safety monitoring mechanism to collect temperature, distance, smoke concentration, human posture and other data in real time. Once risks such as overheating, excessive smoke or patient movement are detected, the arm can be automatically retracted or the smoke removal mode can be switched, and an alarm can be issued or an emergency stop can be executed. Therefore, the present invention has achieved significant improvements in temperature control accuracy, smoke exhaust efficiency and human-machine collaboration, which can meet the multi-level physical therapy needs of hospitals, clinics and families, and has good clinical and industrial value. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a schematic diagram of the overall structure of the moxibustion robot of the present invention;

[0030] Figure 2 This is a schematic diagram of the structure of the body of the moxibustion robot of the present invention;

[0031] Figure 3 It is a structural schematic diagram of the moxibustion module in the present invention;

[0032] Figure 4 It is a schematic diagram of the partial structure of the moxibustion module in the present invention;

[0033] Figure 5 It is a schematic diagram of the split structure of the moxibustion module in the present invention. DETAILED DESCRIPTION

[0034] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments with respect to the moxibustion robot and temperature control method of the present invention.

[0035] In view of the bottlenecks in temperature control accuracy, smoke emission, and human-machine cooperation during the process of traditional Chinese medicine moxibustion physiotherapy, the present invention proposes a comprehensive moxibustion robot solution to achieve efficient, safe, and flexible moxibustion. The present invention aims to achieve the following objectives: First, with the help of two modes of integrated smoke removal of "exhaust fan + primary smoke filter net" and centralized smoke removal of "built-in high-power smoke removal module in the machine body + smoke exhaust pipe", it can be flexibly switched between portable scenarios and large-smoke-volume occasions, improving the smoke exhaust efficiency and adaptability of use; Second, by predicting the change of body surface temperature in advance and dynamically adjusting the distance between the fire head and the skin, overshoot or insufficient temperature can be avoided, thus significantly improving the safety and effect of moxibustion; In addition, the detachable end supports non-stop fire disassembly and quick docking, enabling medical staff to seamlessly switch between manual moxibustion and automatic moxibustion at special acupoints. By integrating sensing data such as temperature, distance, smoke concentration, and human body posture and monitoring in real time, the system can effectively detect risks such as body movement, overheating, and excessive smoke and respond in a timely manner. Its specific structure is as follows:

[0036] As shown in the figure, the moxibustion robot of the present invention includes a body 1 and a robotic arm 2 (a multi-degree-of-freedom robotic arm can be used, such as a 4-axis, 6-axis or 7-axis collaborative robotic arm). A high-power exhaust fan 3 is provided inside the body 1, a filtering unit 4 is provided at the air outlet of the exhaust fan, and a flue gas treatment box 5 is connected to the filtering unit through a section of flue pipe. The said filtering unit can be a multi-stage filter screen (HEPA, activated carbon, etc.) and an odor removal module; of course, it can also include a main power supply for power supply. Thus, powered by the main power supply, it can operate continuously for a long time and is suitable for hospital or multi-bed continuous physiotherapy scenarios. Among them, the robotic arm can be fixed to the body 1 or the side of the treatment bed to adapt to hospital, clinic or home environments; the end of the robotic arm 2 can be mounted with a phototherapy module 6 and a moxibustion module 7. The end moxibustion module 7 includes a moxibustion outer shell 10 having a smoke delivery pipe interface 8 and a primary exhaust port 9, a primary filter element 11 provided inside the moxibustion outer shell, a moxa stick 12 provided at the center of the primary filter element, a moxa stick combustion chamber 13 located at the bottom of the primary filter element and cooperatively installed with the primary filter element, an exhaust fan 14 arranged inside the primary exhaust port and cooperating with the primary filter element, and a heat insulation baffle 15 provided at the bottom end of the moxibustion outer shell. Of course, a battery for powering each electrical component and several sensors (such as smoke and temperature) can also be built inside the moxibustion outer shell. That is to say, the moxibustion module can be equipped with a micro control unit (MCU) and a sensor interface to monitor the combustion temperature, smoke concentration, or communicate with the body controller; when the moxibustion module is disassembled, the control board of the moxibustion module adjusts the fan speed to ensure that the smoke does not diffuse disorderly. The exhaust fan can be powered by the battery, especially a rechargeable battery can be built in, which is sufficient to support the continuous operation of the micro fan and the control circuit in the independent mode. It can also operate independently when the robotic arm is disassembled or away from the main body, forming a local negative pressure adsorption. After inhaling the smoke, preliminary filtration is completed inside the module, and then it is discharged through the smoke delivery pipe interface 8. The fan can provide auxiliary oxygen supply for the thick moxa stick, effectively maintaining the combustion rate and improving the heat release efficiency of the flame head;

[0037] The smoke delivery pipe interface 8 can be connected to the air inlet of the exhaust fan 3 through a smoke delivery pipe 16 (as can be seen from the figure, a series connection port for the smoke delivery pipe 16 is provided at the top of the body), thereby realizing the deep suction and filtration of the smoke. Among them, the said primary filter element is a multi-layer filter screen (activated carbon filter screen, etc.), and the said exhaust fan 14 is a micro fan with adjustable speed (such as a DC brushless fan). The position of the fan 14 corresponds to the primary filter element. During use, even if the end is removed and held by hand, if the flexible pipe is still connected to the body, the centralized smoke removal will still proceed as normal; if the pipe is not connected, it will automatically retreat to the independent fan for smoke exhaust; the body-end decoupling and coupling scheme of the present invention meets the dual requirements of "continuous-fire manual moxibustion + efficient smoke exhaust / oxygen increase".

[0038] Further, a clamping groove can be provided at the end of the robotic arm 2, and a buckle cooperating with the clamping groove is provided on the moxibustion outer shell 10. Through the end moxibustion module 7, it can be clamped at the clamping groove of the robotic arm 2 by the buckle. Of course, electrical / signal interfaces are provided at the clamping groove and the buckle, so that it can work in the state of automatic moxibustion by the robotic arm and can also be removed for manual holding and use.

[0039] Furthermore, a smoke sensor for monitoring smoke concentration information is provided in the body 1. The control signals of the smoke sensor and the exhaust fan are both input to the control unit located in the body. When the upper limit value of the smoke volume or odor concentration (such as using thunder-fire moxibustion or thick moxa sticks) is detected by the smoke sensor, the exhaust fan can be triggered to operate to enhance the exhaust efficiency.

[0040] Furthermore, the moxibustion outer shell 10 is made of heat-insulating materials to protect the internal circuit and prevent the risk of scalding during manual operation; in addition, a heat-insulating metal mesh or a ceramic heat-insulating layer can be provided on the inner wall of the moxibustion outer shell 10 to further prevent the high-temperature fire head from causing harm to the battery, the fan and the hand-held part; the smoke delivery pipe 16 is made of flame-retardant materials and has a certain degree of flexibility to adapt to the movement range of the robotic arm. After the smoke is initially adsorbed, it is deeply exhausted by the exhaust fan through the smoke delivery pipe, achieving secondary smoke removal and odor removal. When the centralized mode is not required, the pipeline connection can be released and returned to the "integrated independent smoke removal" to greatly improve the environmental adaptability. If the fan and the exhaust fan are turned on at the same time, a more significant negative pressure can be generated at the end port, and an appropriate amount of oxygen can be provided when approaching the fire head to ensure the stable fire power of high-heat moxibustion such as thunder-fire moxibustion. The entire robot body will not lose flexibility due to the excessive volume or rigidity of the pipeline, overcoming the defects of the limited movement radius and slow movement of the traditional large-scale pipeline smoking system. It is difficult for those skilled in the art to associate such an integrated design of double-layer smoke exhaust and oxygen-increasing combustion assistance from a simple pipeline smoking scheme. Universal wheels 20 are provided at the bottom of the body. In addition, a handle can be provided on the moxibustion outer shell 10, and an outer heat-insulating strip or heat-dissipating fins can be provided near the hand-held part to prevent high temperature from being conducted to the operator's hand. During manual operation, the fan speed can be set on the moxibustion module through a simple button or knob 21, or remotely adjusted on the upper computer interface; greatly reducing the manual working time and physical burden.

[0041] By configuring a filtering and fan device for each of the robot body and the moxibustion module, and supplementing with a decoupling / coupling flexible pipeline, the present invention achieves a flexible switching scheme between an independent integrated type and centralized deep filtration, and realizes:

[0042] 1. High mobility (detachable end and independent power supply and oxygen increase of the end fan)

[0043] 2. High-efficiency smoke exhaust (double superposition of a high-power filter element of the main unit and the end fan)

[0044] 3. Medical-grade cleanliness (smoke concentration can be monitored in real time and automatically adjusted)

[0045] 4. It is not obvious that it can meet the requirements of portability in small places and high-load smoke exhaust at the same time)

[0046] The present invention designs a detachable moxibustion module at the end of the robot mechanical arm, which supports continuous disassembly and modular expansion, forming a new operation mode of human-machine collaboration. The module can be seamlessly connected or separated from the robot host, allowing medical staff to independently perform moxibustion operations while maintaining the continuous operation of moxa stick combustion and smoke removal functions, overcoming the dilemma of traditional equipment "integrated packaging, disassembly requires shutdown or flameout". In the field of traditional Chinese medicine physiotherapy robots, this innovation completely breaks through the previous idea of ​​simple smoke exhaust arms or single fans at the end, and solves long-standing problems such as large smoke emissions from thick moxa sticks and continuous combustion and oxygenation, with distinct technical progress and broad clinical application prospects.

[0047] The temperature control method of the moxibustion robot of the present invention mainly introduces nonlinear differential equations to perform high-precision modeling of moxibustion combustion heat transfer and body surface heat dissipation, and uses discretization or numerical integration methods to achieve millisecond-level iterative control, thereby achieving significant progress between accuracy and real-time response; this idea overcomes the limitations of single threshold triggering or PID closed-loop hysteresis common in traditional Chinese medicine moxibustion robots, and includes the following steps:

[0048] (A) Establish the nonlinear temperature differential equation

[0049] The existing conventional temperature control method (PID or simple threshold) ignores the coupling effect of moxa stick combustion process on key factors such as distance, ambient temperature, and burner burning rate; the present invention adopts the following model

[0050]

[0051] dt / dT(t): represents the rate of change of temperature T with time t, that is, the derivative of temperature. It describes how temperature changes with time.

[0052] T(t): represents the temperature at time t. This is the variable that the model is trying to predict or control.

[0053] α: It is a constant coefficient, which represents the combined influence of the heat generated by the burning of moxa sticks and factors such as distance and burning rate of the flame. It reflects the contribution of the combustion process to the temperature change.

[0054] h(t): represents the distance between the moxa stick and the target area, which may change with time t. This distance is one of the key factors affecting temperature, because the transfer of heat is related to distance.

[0055] p: is an exponent that represents the degree of influence of distance on temperature change. Different p values will change the sensitivity of distance to temperature change.

[0056] β: is a constant coefficient that represents the degree of influence of the ambient temperature on temperature change. It reflects how the difference between the ambient temperature and the current temperature affects the change in temperature.

[0057] Tenv: represents the ambient temperature, that is, the external ambient temperature during the moxa stick burning process. This is a relatively stable value, but it may change under certain circumstances.

[0058] -β(T(t) - Tenv): This part represents the contribution of the ambient temperature to the temperature change. If the current temperature is higher than the ambient temperature, this part is negative, indicating that heat will be dissipated to the environment; if the current temperature is lower than the ambient temperature, this part is positive, indicating that heat will be absorbed from the environment.

[0059] Where:

[0060] · α represents the heat intensity coefficient of the fire head, which can fluctuate within a certain range depending on the type of moxa stick and the burning state;

[0061] · 1 / h(t) p (p ≥ 2) is used to simulate the heat transfer that decays rapidly as the distance increases;

[0062] · β(T - T env ) follows Newton's law of cooling and is suitable for the physical law of heat dissipation due to the temperature difference between the human body surface and the outside world;

[0063] By distinguishing the heat source heating term and the environmental heat dissipation term, it can more realistically reflect the dynamic changes of temperature with distance and time during the moxibustion process. The traditional methods in this field often simply adjust according to the temperature reading and a fixed proportional coefficient, ignoring the non-linear attenuation and heat dissipation characteristics, and are prone to failure in the case of unstable combustion.

[0064] Can be adapted to a variety of moxa stick types

[0065] For thick moxa sticks such as thunder-fire moxibustion with a fast burning speed and large heat release, α can take a higher value; for ordinary thin moxa sticks, α is relatively reduced; β can also be slightly corrected according to the ambient temperature, humidity, and the patient's skin condition, so that the system can be configured or self-learned and optimized in actual industrial production.

[0066] (B) Iterative control through discretization and numerical integration

[0067] (1) Discretization method

[0068] In actual control, the system executes in a loop with a fixed step size Δt (for example, 5 - 20 ms);

[0069] The finite difference method is adopted, as shown in Equation (2):

[0070]

[0071] Where

[0072] Tn+1: represents the predicted temperature value at time tn+1. This is the temperature at the next time point obtained through iterative calculation.

[0073] Tn: represents the actual temperature value at time tn or the temperature value calculated in the previous iteration step.

[0074] Δt: represents the time step, that is, the time interval between each iteration. In the robot control system, this is usually a fixed value, such as 5 to 20 milliseconds.

[0075] α: is a constant coefficient, which reflects the comprehensive influence of the heat generated by the heat source (such as moxa stick) on factors such as distance and burning rate. This coefficient remains unchanged during the iteration process.

[0076] p: is an exponent, which describes the degree of influence of distance on temperature change. This exponent value is usually obtained through experiments or theoretical calculations and is used to adjust the sensitivity of temperature change to distance.

[0077] β: is a constant coefficient, which represents the degree of influence of the ambient temperature on temperature change. This coefficient reflects how the difference between the ambient temperature and the current temperature affects the temperature change.

[0078] Tenv: represents the ambient temperature, that is, the external ambient temperature where the heat source is located. This value is usually considered constant during the iteration process.

[0079] hn: represents the distance between the end of the robotic arm (or heat source) and the body surface at time tn. This distance changes with time and is one of the key factors affecting temperature change.

[0080] By iterating step by step, the temperature increment at the "next moment" can be predicted in a very short time. Compared with the traditional PID repeated parameter tuning or lag control, the instant iteration + look-ahead method of the present invention has higher sensitivity and control accuracy.

[0081] (2) Higher-order numerical integration is optional

[0082] If Δt increases or higher accuracy is required, the present invention can also use more stable medium / high-order methods (such as Runge-Kutta, linear multi-step methods, etc.) to solve the differential equation.

[0083] (3) Combination with robotic arm motion planning

[0084] After each discrete step updates the distance / temperature, the system immediately passes parameters such as the new desired distance Δh to the trajectory generation module, and actually moves the end position of the robotic arm through inverse kinematics and joint interpolation.

[0085] This tight coupling of temperature control and motion is rare in previous moxibustion robots. Most of them only perform simple retraction actions when the temperature exceeds the limit, while the present invention seamlessly integrates differential equations with trajectory control to achieve millisecond-level response, improving clinical safety and moxibustion effect.

[0086] In summary, in the "Nonlinear temperature equation and discretization / numerical integration" section, the present invention actively predicts the temperature change trend through a nonlinear differential equation based on the distance between the moxibustion head and the body surface, and uses high-speed numerical iteration in robot control systems such as ROS to link the end position of the robotic arm, successfully solving problems such as temperature control lag and frequent overshoot in traditional Chinese medicine moxibustion robots.

[0087] Finally, for the moxibustion manipulation motion trajectory and algorithm,

[0088] In the present invention, for different moxibustion manipulation methods (such as "pecking moxibustion" and "rotary moxibustion") required during traditional Chinese medicine moxibustion, we introduce a series of three-dimensional motion trajectory formulas to describe the position of the end of the robotic arm (TCP) changing with time. Combining correction amounts such as Δh(t) and Δr(t) given by the nonlinear temperature control model, the trajectory can be automatically adjusted according to the body surface temperature and the temperature rise rate. Thus, it not only retains the operation characteristics of traditional moxibustion methods but also avoids the risk of scalding caused by temperature overshoot or body movement. This part will elaborate in more detail on the formulas and their discrete implementation, and explain how to implement these formulas into the executable trajectory of robot motion planning (such as ROS) to meet industrial and clinical feasibility.

[0089] A. Overview of Discretization and Numerical Integration

[0090] Numerical decoupling of the temperature equation: In the temperature control module of the present invention, we use a differential equation (or a similar function) to describe the rate of change of the body surface temperature T with time:

[0091]

[0092] T(t): Body surface temperature (a function that changes with time)

[0093] t: Time, representing the time variable during the moxibustion process.

[0094] dT(t) / dt: Represents the rate of change of the body surface temperature T with time t. This is the left side of the differential equation, describing how the temperature changes with time.

[0095] F(h(t)): A function of the distance h(t), representing the impact of the distance between the heat source and the body surface on temperature change.

[0096] h(t): Represents the distance between the heat source and the body surface at time t. This distance changes with time and is one of the key factors affecting temperature change.

[0097] γ (gamma): A constant coefficient representing the degree of influence of the ambient temperature on temperature change. This coefficient reflects how the difference between the ambient temperature and the current temperature affects the rate of temperature change. It is similar to the reciprocal of the heat conduction coefficient or the heat loss coefficient and is used to adjust the sensitivity of temperature change to the ambient temperature.

[0098] T_{env}: The ambient temperature, i.e., the background temperature in the moxibustion environment.

[0099] -γ(T(t) - T_{env}): This part represents the contribution of the ambient temperature to the rate of temperature change. If the current temperature is higher than the ambient temperature, this part is negative, indicating that heat will be dissipated to the environment, resulting in a decrease in the rate of temperature change; if the current temperature is lower than the ambient temperature, this part is positive, indicating that heat will be absorbed from the environment, resulting in an increase in the rate of temperature change.

[0100] Or a more complex logarithmic / exponential model can be adopted, in which the temperature increment decays according to an inverse power law with the change of the distance h and releases heat to the ambient temperature T according to Newton's law of cooling. env Discretization method: Use the finite difference (Euler method) or high-order numerical integration (such as Runge-Kutta) to iterate T(t) at each time step Δt to obtain T n+1 ≈T n +Δt×…, and thus estimate the time required to reach the target temperature T target and then output Δh n or Δr n for trajectory correction.

[0101] Tight coupling with trajectory control: When the robotic arm performs "pecking moxibustion" or "circular moxibustion", the temperature control module calculates whether to move the end up / down (Δh), increase / decrease the circular radius (Δr, etc.) according to the latest temperature iteration result in each cycle; superimpose these correction values on the originally fixed sine or circular arc trajectory, so that the trajectory can automatically make fine adjustments following the temperature change. This predictive + discrete iteration method cannot be easily thought of by those of ordinary skill in the art, reflecting the creativity of the present invention in the combustion / moxibustion scenario.

[0102] B. Basic three-dimensional trajectory formula

[0103] Without considering the dynamic correction of temperature control, the present invention first provides the following typical trajectory patterns, which respectively correspond to different traditional Chinese medicine moxibustion methods:

[0104] (1) Fixed-point pecking moxibustion (vertical sine reciprocation)

[0105] P peck (t) = (x0, y0, z0 + H) + (0, 0, A sin(ωt)), (1)

[0106] · (x0, y0, z0) is the reference position (acupoint coordinates), and H is the reference height. A is the reciprocating amplitude, and ω = 2πf is the angular frequency.

[0107] · The trajectory performs periodic up-and-down vibrations in the Z direction to simulate the "pecking" action.

[0108] P(t): The desired pose of the end of the robotic arm (a function of time), representing the position coordinates of the end in three-dimensional space, usually composed of ((x(t), y(t), z(t))).

[0109] (x_0, y_0, z_0): The reference position, representing the fixed coordinates of the moxibustion acupoint.

[0110] H: The reference height, representing the average distance between the fire head and the body surface.

[0111] A: The reciprocating amplitude, representing the amplitude of the up-and-down oscillation of the pecking moxibustion.

[0112] ω (omega): The angular frequency, representing the angular velocity of the oscillation, defined as ω = 2πf

[0113] f: The frequency, representing the frequency of the reciprocating movement of the pecking moxibustion.

[0114] t: The time, representing the time variable during the moxibustion process.

[0115] sin(ωt): The sine function, used to generate periodic up-and-down movement.

[0116] (0, 0, A sin(ωt)): The oscillation displacement vector, representing the displacement changes of the end of the robotic arm in the X, Y, and Z directions.

[0117] 1) Circular moxibustion (horizontal arc or circle)

[0118] P circ (t) = (x0 + r cos(ωt), y0 + r sin(ωt), z0 + H), (2)

[0119] · r is the fixed radius; if a Z-direction periodic term or an inclined plane angle is added, a helix or multi-degree-of-freedom circular motion can be formed.

[0120] Pcirc(t): represents the position coordinates at time t, which is a function of time t and gives the coordinates of the points that change with time during the cyclotron process

[0121] x0: represents the coordinate of the center of the gyration in the x-axis direction. This is the x-coordinate of the center of the gyration trajectory.

[0122] r: represents the fixed radius of the gyration. This is the distance from the center of the gyration to any point on the trajectory.

[0123] cos(ωt): Cosine function, where ω is the angular frequency (the angle rotated per unit time) and t is time. This function describes the displacement change in the x direction over time, making the trajectory present a periodic sine wave shape in the x direction (but after multiplying with r here, it forms part of a circular motion)

[0124] y0: represents the coordinate of the center of the spiral in the y-axis direction. This is the y coordinate of the center of the spiral trajectory.

[0125] sin(ωt): Sine function, similar to cosine function, but with a phase difference of 2π (i.e. 90 degrees). This function describes the change in displacement in the y direction over time, and together with cos(ωt), the two combine to form circular motion.

[0126] z0: represents the base height (or average height) of the loop track in the z-axis direction. This is the base value of the z coordinates of all points in the loop track.

[0127] H: represents a fixed offset in the z direction. This value does not change with time, and it makes the entire spiral trajectory have a uniform increase or decrease in the z direction. t: represents time. This is the independent variable in the formula, representing the time point at any moment in the spiral process.

[0128] (2) Physical intuition and industrial feasibility

[0129] By setting the round-trip amplitude d, turning radius r, frequency f, etc., common moxibustion movements can be generated. It is simple and easy to use on industrial robots, which is conducive to large-scale production and promotion in medical scenarios.

[0130] C. Dynamic correction: variable height and variable radius

[0131] In order to cope with temperature overshoot or personalized moxibustion requirements, the present invention adds additional correction items to the above basic trajectory, corresponding to the discrete iteration results of the temperature control module:

[0132] 1. Vertical height correction Δh(t)

[0133] definition:

[0134] z0 + H → z0 + H + Δh(t), (3)

[0135] Δh(t) is obtained by iterating the temperature equation. In each discrete period, if it is predicted that the target temperature Ttarget will be reached or exceeded in a short time, Δh(t) increases, causing the end to move upward as a whole; if the temperature rise is insufficient, Δh(t) decreases or becomes negative.

[0136] z0: Represents the reference height (or initial height, average height) of the circular motion trajectory in the z-axis direction. This is the reference value of the z-coordinate of all points on the circular motion trajectory when no vertical height correction is performed.

[0137] H: Represents a fixed offset in the z direction. This value does not change with time, and it causes the entire circular motion trajectory to have a unified elevation or depression in the z direction, which is a fixed adjustment based on the reference height z0.

[0138] Δh(t): Represents the vertical height correction amount that changes with time. This is a function of time t and is used to dynamically adjust the vertical height of the end according to certain conditions (such as temperature).

[0139] z0 + H: Represents the height of the end in the z-axis direction before vertical height correction.

[0140] z0 + H + Δh(t): Represents the height of the end in the z-axis direction after vertical height correction. This is the height of the end during the actual circular moxibustion process.

[0141] Ttarget: Represents the target temperature. This is the temperature value that is desired to be reached or maintained during the circular moxibustion process.

[0142] 2. Dynamic radius correction Δr(t)

[0143] In circular moxibustion:

[0144] r → r + Δr(t) (4)

[0145] For areas with a large body surface area or areas that need to be evenly covered, Δr(t) can be increased or decreased periodically, or determined according to the gradual change of temperature, to avoid overheating or insufficient moxibustion caused by a single radius.

[0146] r: Represents the initial radius or reference radius of circular moxibustion. This is the distance from any point on the circular motion trajectory to the center of the circle when no dynamic radius correction is performed.

[0147] Δr(t): Represents the radius correction amount that changes with time. This is a function of time t and is used to dynamically adjust the radius of the circular motion according to certain conditions (such as temperature, the body surface area of the treatment area, or the degree of even coverage required).

[0148] r + Δr(t): represents the actual gyroradius after dynamic radius correction. This is the radius used during the actual moxibustion process.

[0149] 3. Multiple Combinations

[0150] Combining (3) and (4), time-varying control in the vertical (height) and planar (radius) directions can be carried out simultaneously. For example:

[0151] P spiral-dyn (t) = (x0 + [r + Δr(t)]cos(ωt), y0 + [r + Δr(t)]sin(ωt), z0 + H + Δh(t)) can construct a spiral motion with a variable radius and make real-time fine-tuning in the Z direction. This three-dimensional variable trajectory plays a key role in the coverage and temperature uniformity of the moxibustion site, far exceeding traditional operations with a fixed radius or height.

[0152] Pspiral-dyn(t): represents the position coordinates of the three-dimensional dynamic spiral trajectory at time t. This is a function of time t, giving the coordinate points that change with time during the moxibustion process, including dynamic adjustments of the radius and height.

[0153] x0: represents the coordinate of the center of gyration in the x-axis direction. This is the x coordinate of the center of the gyration trajectory and is one of the reference points of the trajectory.

[0154] r: represents the initial radius or reference radius of the moxibustion. This is the distance from any point on the gyration trajectory to the center of gyration when there is no dynamic radius correction.

[0155] Δr(t): represents the radius correction amount that changes with time. This is a function of time t, used to dynamically adjust the radius of gyration according to certain conditions (such as temperature, the body surface area of the treatment area, or the degree of uniform coverage required).

[0156] ω: represents the angular frequency, that is, the angle turned per unit time. This parameter determines the speed or period of the moxibustion.

[0157] t: represents time. This is the independent variable in the formula, representing the time point at any moment during the moxibustion process.

[0158] y0: represents the coordinate of the center of gyration in the y-axis direction. This is the y coordinate of the center of the gyration trajectory and, together with x0, determines the reference point of the trajectory.

[0159] z0: represents the reference height (or initial height) of the gyration trajectory in the z-axis direction. This is the reference value of the z coordinates of all points on the gyration trajectory when there is no vertical height correction.

[0160] H: Represents a fixed offset in the z - direction. This value does not change with time and causes the entire gyration trajectory to have a unified elevation or depression in the z - direction. Δh(t): Represents the vertical height correction that changes with time. This is a function of time t and is used to dynamically adjust the vertical height of the end - effector according to certain conditions (such as temperature).

[0161] D. Algorithm Implementation: From Formulas to Robot Motion Planning

[0162] (1) Trajectory Generation and Inverse Kinematics

[0163] a. On ROS or an equivalent system, the present invention sets up a "trajectory generation node" that periodically reads correction values such as Δh(t) and Δr(t) and embeds them into formulas such as (1) - (4).

[0164] b. Obtain the desired end - effector pose P(t), then call an inverse kinematics solver (MoveIt, KDL, etc.) to obtain the feasible joint angles q(t) and perform velocity / acceleration interpolation, and execute it on the robotic arm.

[0165] (2) Discretization Period: The typical control step size can be set to 10 - 50 ms. The temperature module uses numerical integration for discrete iteration to calculate new Δh / Δr. The trajectory node substitutes these new parameters into the trajectory equation to obtain the end - effector coordinates at the next moment. The robotic arm actuator performs interpolation motion.

[0166] (3) Safety Trigger: If the smoke sensor or human body sensor reports an anomaly, the system can interrupt the interpolation at any discrete period, causing the robotic arm to return to a safe pose (such as raising Z or moving to one side), minimizing the risk of scalding or collision. This ensures the safety and reliability required for medical devices and complements the "variable radius / height" algorithm.

[0167] In addition, the present invention not only innovates in temperature control and manipulation trajectories, but also configures a multi - modal data fusion and automated safety monitoring mechanism for the high - safety requirements of the traditional Chinese medicine moxibustion scenario. This mechanism makes full use of information such as human body posture sensors, smoke sensors, temperature and distance detection, etc., to provide the moxibustion robot with real - time closed - loop dynamic scheduling capabilities, ensuring that the treatment process is controllable, effective and compliant with medical device regulations. The specific structure is as follows:

[0168] (1) Infrared array - type temperature sensors, laser / ultrasonic rangefinders, etc. can be built - in around the moxibustion module or the fuselage to accurately obtain the body surface temperature distribution or the distance between the fire head and the skin. In the non - linear temperature control model, this sensing information will be used as key inputs to trigger differential equation iteration or numerical integration. Different from the traditional scheme that only uses single - point temperature measurement, the present invention can collect multi - point temperatures, identify local overheating and perform local retraction such as Δh, etc., with a higher safety margin.

[0169] (2) A smoke sensor is installed on the moxibustion module to detect the smoke concentration. When the concentration exceeds the set threshold (e.g., greater than the ppm standard), the host computer immediately issues an alarm and can automatically switch to the centralized smoke removal mode, or increase the fan speed to strengthen smoke exhaust. Compared with similar devices that only have an external large smoking machine and no concentration detection, the multi-stage smoke exhaust + sensing fusion of the present invention greatly improves the smoke removal efficiency and the medical staff's comfort level.

[0170] (3) Corresponding human body posture / body movement sensors are set up. Specifically, methods such as a visual depth camera, an infrared stereo sensor, or a somatosensory radar can be used to cover the moxibustion area. Once it detects that the patient has a large body movement (such as turning over or local movement), the system can issue a prompt and automatically pause the movement of the robotic arm or retreat to a safe distance to prevent scalding or collision. This active body movement detection is not yet popular in traditional Chinese medicine moxibustion robots, reflecting the in-depth consideration of the safety requirements of the present invention.

[0171] To improve reliability, temperature / smoke sensors can be installed on the end moxibustion module and the fuselage respectively to form a dual redundant detection. Once a certain sensor fails or the data is abnormal, the system can still maintain the basic safety monitoring function.

[0172] The fusion strategy method is as follows:

[0173] ① Transmit the temperature and distance data to the non-linear predictive temperature control to update the trajectory correction values such as Δh and Δr;

[0174] ② Link the smoke concentration information with the dual-mode smoke removal, and automatically switch to the centralized smoke removal when necessary;

[0175] ③ Combine the body movement detection result with the detachable end coordination or the trajectory dynamic correction. If the patient or the module has left the expected moxibustion area, stop the movement of the robotic arm or trigger an alarm.

[0176] The functions of its main components are as follows:

[0177] Moxa stick combustion chamber: Place the moxa stick and maintain stable combustion; Anti-ash metal mesh: Block the ashes from falling onto the fan or the circuit area; Exhaust port & smoke delivery pipe interface: Can either directly discharge the processed gas or connect a flexible pipe to send the smoke into the fuselage for secondary filtration; Fan & replaceable smoke filter module: Achieve primary filtration and increase oxygen to assist combustion, and the filter element can be quickly replaced; The fan and the primary filter element inside the end moxibustion module can already complete the primary smoke treatment, and the clean gas can be directly discharged through the exhaust port; Centralized smoke removal: In occasions with a large amount of smoke or strong odor, connect with the end moxibustion through the flexible pipeline of the fuselage, and introduce the flue gas into the high-power fan + multi-stage filter element of the fuselage for deep filtration.

[0178] Temperature prediction and motion control implementation steps

[0179] 1. Real-time acquisition of temperature and distance

[0180] An infrared temperature sensor, a multi-point temperature measurement array or a laser ranging device is arranged near the end moxibustion module or the body, which is used to measure the skin surface temperature and the distance between the flame head and the skin.

[0181] 2. Implementation of non-linear predictive temperature control algorithm

[0182] The upper computer performs iterative calculations on the change rate of the temperature T according to the pre-established coupling model of heat source heating and body surface heat dissipation; within each control cycle, the short-term future temperature trend is predicted based on numerical integration or the Runge-Kutta method.

[0183] If the prediction result shows that the temperature will exceed the set threshold, the upper computer immediately calculates and outputs the motion increment (Δh) of "raising / retreating", and commands the robotic arm to increase the end height or reduce the moxibustion radius, so as to pre-reduce the heat transfer from the flame head to the skin before the temperature overshoots.

[0184] For moxa sticks with different thicknesses and combustion speed differences, the model parameters (such as α, β) can be adjusted in the control software to avoid temperature control deviations.

[0185] 3. Trajectory generation and superposition correction

[0186] Based on the traditional "pecking moxibustion" or "circular moxibustion" trajectory, the system periodically superimposes correction amounts such as Δh and Δr output by the temperature control algorithm on the original fixed reciprocating or circular motion equation, so that the moxibustion trajectory is dynamically adjusted according to the actual temperature.

[0187] By calling the robotic arm motion controller (inverse kinematics solution and velocity interpolation), the joint angle command is updated in real time; once a risk or over-limit temperature is detected, an emergency stop or retraction operation is immediately executed.

[0188] Dual-mode smoke removal implementation process

[0189] 1. Independent integrated smoke exhaust

[0190] By default, the micro fan and the preliminary filter element of the end moxibustion module form a local negative pressure to preliminarily filter the smoke generated during moxibustion; the filtered gas is discharged through the exhaust port.

[0191] In this mode, there is no need to connect the body pipeline, which is suitable for portable scenarios such as families or small clinics, reducing the system occupancy space and assembly burden.

[0192] 2. Centralized deep filtration

[0193] If the smoke sensor detects a continuous increase in concentration or a large-smoke moxa stick such as thunder-fire moxibustion is used, the upper computer can prompt or automatically enter the centralized mode.

[0194] The flexible pipe is connected to the end moxibustion. After the smoke is initially inhaled by the fan, it is further introduced into the high-power filtering unit of the fuselage through the pipe, thereby realizing secondary smoke exhaust and odor removal.

[0195] In the centralized mode, the large fan and the small fan at the end can work together. On the one hand, it accelerates the smoke exhaust, and on the other hand, it provides a certain amount of oxygen-increasing airflow for the moxa stick to improve the combustion efficiency.

[0196] It can be seen that the moxibustion robot of the present invention mainly includes: a fuselage, which is internally provided with a high-power filtering unit and a main control unit. The main control unit is electrically connected to an external power supply and a multi-degree-of-freedom robotic arm; a multi-degree-of-freedom robotic arm, which is installed on the fuselage or its base and is used for spatial positioning and moxibustion application actions; a moxibustion module, which is detachably installed at the end effector of the robotic arm through a snap or slot structure. The end moxibustion module is internally provided with a micro fan and a primary filtering unit, which are used for primary filtering of moxibustion smoke and maintaining the combustion of the moxa stick; a multi-modal sensor, which at least includes a temperature sensor, a distance sensor, and a smoke sensor, and is used for collecting information such as body surface temperature, the distance between the fire head and the skin, and smoke concentration; a non-linear predictive temperature control system. The main control unit predicts the moxa stick combustion and body surface heat dissipation processes using non-linear differential equations based on the collected multi-modal sensing data, and real-time adjusts the position or movement trajectory of the end of the robotic arm to keep the body surface temperature within the safe threshold range;

[0197] The moxibustion module is internally provided with an independent battery, which is used to continuously supply power to the micro fan even after being separated from the robotic arm, so that the moxa stick combustion and smoke removal functions are not interrupted; the moxibustion module has an anti-scald structure and a heat insulation layer, and an anti-ash net is provided outside the moxa stick combustion chamber to prevent the moxa ash or high-temperature fire head from directly contacting the internal electronic devices or the operator's hand; the high-power filtering unit in the fuselage includes multiple filter meshes and a fan. When the detected smoke concentration exceeds the preset threshold, the exhaust air speed is automatically switched or enhanced, and together with the fan of the end moxibustion module, it forms secondary smoke removal. The multi-modal sensor also includes a human body posture detection sensor or an infrared / depth camera, which is used to monitor the patient's body movement or posture change in real time. Once a large displacement is detected, the main control unit will make the robotic arm automatically retreat to a safe distance or pause the moxibustion application action; the robotic arm performs discrete iterative control according to the temperature change trend output by the non-linear predictive temperature control system. If the predicted temperature is about to exceed the safe threshold, the end moxibustion module is lifted or moved backward in advance to avoid overshoot; the non-linear predictive temperature control system calculates the body surface temperature and combustion heat using the method of discretization or numerical integration, specifically including:

[0198] In each control cycle, the body surface temperature is estimated according to the data obtained by the temperature and distance sensors; when the predicted temperature will exceed the target range, a correction amount Δh or Δr is generated and superimposed and corrected in the movement trajectory of the robotic arm in the next cycle.

[0199] The non-linear differential equation combines parameters such as the combustion rate coefficient, the environmental heat dissipation coefficient, and the body surface characteristics, and can adaptively adjust coefficients such as α and β according to different moxa stick types or environmental conditions to ensure the accuracy of moxibustion temperature control. When the robotic arm executes the trajectories of different traditional Chinese medicine moxibustion methods such as "pecking moxibustion" and "rotary moxibustion", it can automatically inject the pose correction amount output by the non-linear predictive temperature control system to achieve real-time adjustment of the temperature-driven trajectory; the end moxibustion module is equipped with a quick docking and positioning device, and there is no need to turn off the moxa stick combustion or stop the fan during disassembly and assembly. The operator can complete the switching between manual moxibustion and automatic moxibustion within a short time. In addition, once the system detects that the value of any temperature sensor or smoke sensor reaches the limit threshold, it immediately triggers the emergency stop of the robotic arm and sends an alarm message to the operator, while keeping the end fan working to avoid soot accumulation. A display or indicator light component is set on the moxibustion module, which can display the status information such as the current temperature, smoke concentration, and fan speed in real time, facilitating medical staff to immediately master the moxibustion parameters during hand-held operation.

[0200] Finally, the basic control method of the present invention is as follows:

[0201] This control method includes:

[0202] (1) Data acquisition: Periodically obtain sensor data such as temperature, distance, smoke concentration, and human body posture;

[0203] (2) Prediction calculation: Input the temperature and distance data into the non-linear temperature model for numerical integration or discrete iteration to predict the change in body surface temperature after several step lengths;

[0204] (3) Trajectory correction: Calculate the correction amount of the height or trajectory radius between the end of the robotic arm and the skin according to the comparison result of the predicted temperature and the safety threshold;

[0205] (4) Motion execution: Superimpose the correction amount onto the established motion trajectory of traditional Chinese medicine moxibustion techniques, and through inverse kinematics solution and velocity interpolation, achieve real-time dynamic moxibustion;

[0206] (5) Smoke removal control: Automatically switch between the "integrated end smoke exhaust" or "body centralized filtration" mode according to the smoke concentration data, and start the high-power filtration unit when a large amount of smoke is identified;

[0207] (6) Safety monitoring: If it is detected that the patient's body movement, temperature, or smoke exceeds the limit, immediately execute the avoidance or emergency stop of the robotic arm, and keep the end fan running until the hazard is eliminated;

[0208] (7) Detachable operation: The end moxibustion module can be detached for manual moxibustion without extinguishing the fire, and after completion, it is re-docked to the robotic arm, and the system automatically resumes the original trajectory and temperature control logic.

[0209] In addition, during the non-linear temperature prediction in Step 2, different types of moxibustion consumables such as thunder-fire moxibustion and thin moxa sticks are adapted through customized combustion rate coefficients and heat dissipation parameters to achieve more precise temperature control. When performing the smoke removal control in Step 5, the air volumes of the end small fan and the main body fan of the machine are automatically adjusted according to the concentration value of the smoke sensor, so as to dynamically switch between the portable mode and the high-load mode, and give a replacement or maintenance prompt when the filter element is blocked or the temperature is too high.

[0210] Certainly, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by those skilled in the art within the scope of the essence of the present invention should also fall within the protection scope of the present invention.

Claims

1. A moxibustion robot, comprising a body (1) and a mechanical arm (2), characterized in that: The machine body (1) is provided with a fan (3), a filter unit (4) arranged at the air outlet of the fan, and a smoke treatment box (5) connected to the filter unit through a smoke pipe. The end of the mechanical arm (2) can be mounted with a phototherapy module (6) and a moxibustion module (7). The moxibustion module (7) includes a moxibustion shell (10) having a smoke delivery pipe interface (8) and a primary exhaust port (9), a primary filter element (11) arranged in the moxibustion shell, a moxa stick (12) arranged in the center of the primary filter element, a moxa stick combustion chamber (13) located at the bottom of the primary filter element and installed in cooperation with the primary filter element, an exhaust fan (14) arranged in the primary exhaust port and in cooperation with the primary filter element, and an anti-scalding baffle (15) arranged at the bottom of the moxibustion shell. The smoke delivery pipe interface (8) can be connected to the air inlet of the fan (3) through a smoke delivery pipe (16) to achieve deep suction and filtration of smoke.

2. The moxibustion robot according to claim 1, characterized in that: The end of the mechanical arm (2) is provided with a slot, and the moxibustion housing (10) is provided with a buckle that matches the slot, so that the end moxibustion module (7) can be buckled in the slot of the mechanical arm (2) through the buckle.

3. The moxibustion robot according to claim 1 or 2, characterized in that: The moxibustion shell (10) is made of heat-insulating material.

4. The moxibustion robot according to claim 3, characterized in that: The primary filter element is a multi-layer filter screen, and the exhaust fan (14) is a DC brushless fan.

5. The moxibustion robot according to claim 1, 2 or 4, characterized in that: The machine body (1) is provided with a smoke sensor for monitoring smoke concentration information. The control signals of the smoke sensor and the fan are input into a control unit located in the machine body. When the smoke sensor detects an upper limit value of the smoke volume or odor concentration, the fan can be triggered to operate to enhance smoke exhaust efficiency.

6. The moxibustion robot according to claim 5, characterized in that: The smoke delivery duct (16) is made of flame-retardant material.

7. The moxibustion robot according to claim 1, 2, 4 or 6, characterized in that: The bottom of the machine body is provided with universal wheels (20).

8. A temperature control method for a moxibustion robot according to any one of claims 1 to 7, characterized in that: The following steps are involved: (A) Establish the nonlinear temperature differential equation The following model is used in: α represents the thermal intensity coefficient of the fire head, which can fluctuate within a certain range depending on the type of moxa stick and the burning state; 1 / h(t) p (p ≥ 2) is used to simulate heat transfer that decays sharply with increasing distance; β(TT env ) Comply with Newton's law of cooling, which is suitable for the physical law of heat dissipation due to the temperature difference between the human body surface and the outside world; (B) Iterative control through discretization In actual control, the system executes cyclically with a fixed step length Δt (e.g., 5 to 20 ms); oUse the finite difference method, as shown in formula (2): where h n Indicates that at time t n The distance between the end of the robotic arm and the body surface. By iterating step by step, the temperature increment of the "next moment" can be predicted in a very short time.

9. The temperature control method of the moxibustion robot according to claim 8, characterized in that: If Δt increases, use intermediate / higher order methods to solve the differential equations.

10. The temperature control method of the moxibustion robot according to claim 9, characterized in that: After updating the distance / temperature at each discrete step, the system immediately passes the new expected distance Δh parameter to the trajectory generation module, which actually moves the end position of the robot arm through inverse kinematics and joint interpolation.

Images