An automatic following type medical infusion system and method

Abstract

The application relates to an automatic following type medical infusion system and method, and relates to the technical field of medical infusion. The system comprises a binding strength module, a tracking accuracy module, a path planning module, a speed control module and an obstacle avoidance decision module. The binding strength module is used for obtaining dynamically updated binding strength of a patient and an infusion device; the tracking accuracy module is used for obtaining tracking accuracy of the patient after being calibrated by the binding strength; the path planning module is used for obtaining an optimal path of the infusion device movement; the speed control module is used for obtaining a current speed of the infusion device subjected to adaptive control according to feedback values of the optimal path and the tracking accuracy; the obstacle avoidance decision module is used for obtaining a threat degree of an obstacle according to the current speed and obstacle scanning data of a three-dimensional laser radar, and determining an obstacle avoidance decision value of the infusion device based on the threat; and the automatic following module is used for obtaining a motion control amount of the infusion device according to a cooperative feedback of the optimal path, the current speed and the obstacle avoidance decision value of the infusion device. The system can realize accurate automatic following of the infusion device to the patient for infusion.

Description

Technical Field

[0001] This invention relates to the field of intelligent sensors, and more particularly to the field of medical infusion technology, specifically to an automatic follow-up medical infusion system, method, electronic device, and non-transitory computer-readable storage medium. Background Technology

[0002] Currently, during intravenous infusions in hospitals, patients typically need to carry an IV stand or cart to move around. Whether going for examinations, using the restroom, or simply taking a short break, the IV cart must be manually pushed or pulled. Existing medical infusion devices mainly consist of an IV stand with casters, requiring manual movement by the patient or their caregiver.

[0003] However, patients need to control the IV cart with one hand, which restricts their freedom of movement and increases the risk of falls or medical accidents. On the other hand, in the crowded and space-constrained environment of a hospital, manually pushing the cart is not only inconvenient but may also obstruct the passage of others. Summary of the Invention

[0004] This invention addresses the technical problems existing in the prior art by providing an automatic following medical infusion system and method that enables infusion devices to accurately and automatically follow patients during infusion.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] This invention provides an automatic following medical infusion system, the system comprising:

[0007] The binding strength module is used to obtain the dynamically updated binding strength between the patient and the infusion device based on the matching results of the RFID tag signal worn by the patient and the identity database.

[0008] The tracking accuracy module is used to obtain the tracking accuracy of the patient calibrated by the binding strength based on the binding strength and the distance and speed data collected in real time by multiple sensors.

[0009] The path planning module is used to obtain the optimal path for the infusion device to move based on the tracking accuracy and the spatial topology data obtained by the environmental perception module.

[0010] A speed control module is used to obtain the current speed of the infusion device under adaptive control based on the feedback value of the optimal path and the tracking accuracy.

[0011] The obstacle avoidance decision module is used to obtain the threat level of the obstacle based on the current speed and the obstacle scanning data of the three-dimensional lidar, and to determine the obstacle avoidance decision value of the infusion device based on the threat.

[0012] The automatic following module is used to obtain the motion control quantity of the infusion device based on the collaborative feedback of the optimal path, current speed and obstacle avoidance decision value of the infusion device.

[0013] Furthermore, the binding strength module is also used for:

[0014] Obtain an initial binding coefficient that represents the initial connection strength calibration value between the patient and the infusion device;

[0015] Construct a time decay term based on the time decay factor and the current time.

[0016] Obtain the periodic fluctuation factor used to simulate the periodic fluctuations of the bound signal;

[0017] The initial binding coefficient is processed based on the integral value of the binding signal strength at historical moments, the time decay term, and the periodic fluctuation factor to obtain the binding strength.

[0018] Furthermore, the tracking accuracy module is also used for:

[0019] Dynamic weights are configured for different sensors to obtain the sensor weights for each sensor.

[0020] Based on the sensor weights of each sensor, the relative distance and relative speed measured by each sensor are fused together;

[0021] The reliability of the fused data is adjusted based on the reliability index of each sensor, and the final accuracy is calibrated by combining the real-time binding strength to obtain the tracking accuracy.

[0022] Furthermore, the path planning module is also used for:

[0023] Based on the path curvature function and tracking accuracy of each path, a control overhead term for positioning and tracking accuracy is constructed;

[0024] Calculate the square of the rate of change of the angle of the infusion device traveling along each of the aforementioned paths;

[0025] Calculate the unit energy consumption of the infusion device under each of the path controls;

[0026] The optimal path is obtained by combining the control overhead, the square of the angle change rate, and the unit energy consumption, and then performing differentiation.

[0027] Furthermore, the speed control module is also used for:

[0028] The tracking accuracy is adjusted based on the velocity response coefficient to obtain the adjusted tracking accuracy.

[0029] Calculate the rate of change of the binding strength over time;

[0030] Calculate the cumulative path cost based on the path impact factor and the optimal path at each historical moment;

[0031] The current speed is determined based on the adjusted tracking accuracy, the time rate of change of the binding strength, and the cumulative path cost.

[0032] Furthermore, the obstacle avoidance decision module is also used for:

[0033] Obtain the threat level of each obstacle;

[0034] The threat level of each obstacle is calculated based on the distance and azimuth angle between the infusion device and each obstacle.

[0035] Based on the threat level and threat intensity of each obstacle, determine the obstacle avoidance sensitivity of each obstacle, and then merge the obstacle avoidance sensitivities of all obstacles in the path.

[0036] The obstacle avoidance decision value is obtained by dynamically adjusting the fused obstacle avoidance sensitivity based on the current speed of the infusion device.

[0037] Furthermore, the automatic following module is also used for:

[0038] Calculate the rate of change of the optimal path over time;

[0039] Based on the rate of change of the optimal path over time, the current speed, and the obstacle avoidance decision value, and in conjunction with the control gain coefficients corresponding to each of the above parameters, the motion control quantity is determined.

[0040] Furthermore, to achieve the above objectives, the present invention also proposes an automatic following medical infusion method, the method comprising:

[0041] Based on the matching results between the RFID tag signal worn by the patient and the identity database, the binding strength between the patient and the infusion device is dynamically updated.

[0042] Based on the binding strength and the distance and speed data collected in real time by multiple sensors, the tracking accuracy of the patient calibrated by the binding strength is obtained;

[0043] Based on the tracking accuracy and the spatial topology data obtained by the environmental perception module, the optimal path for the movement of the infusion device is obtained;

[0044] Based on the feedback value of the optimal path and the tracking accuracy, the current speed of the adaptively controlled infusion device is obtained;

[0045] Based on the current speed and obstacle scanning data from the three-dimensional lidar, the threat level of the obstacle is obtained, and the obstacle avoidance decision value of the infusion device is determined based on the threat.

[0046] The motion control quantity of the infusion device is obtained based on the collaborative feedback of the optimal path, current speed, and obstacle avoidance decision value of the infusion device.

[0047] In addition, to achieve the above objectives, the present invention also proposes an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby realizing an automatic follow-up medical infusion method as described above.

[0048] Furthermore, to achieve the above objectives, the present invention also proposes a non-transitory computer-readable storage medium storing a computer software program, which, when executed by a processor, implements an automatic follow-up medical infusion method as described above.

[0049] The beneficial effects of this invention are:

[0050] (1) Traditional infusion methods often require patients to be confined to a static infusion stand, while this solution enables dynamic movement of the infusion equipment through an automatic following system, allowing patients to walk freely, go up and down for examinations, use the toilet and other daily activities during the infusion process, which greatly improves the hospitalization experience.

[0051] (2) By integrating multiple sensors such as ultrasound, infrared, and vision, the system has high-precision target recognition and tracking capabilities. The mathematical model introduces the binding strength B(t), which can dynamically evaluate the stability of the patient-device association and ensure that the system continues to follow without deviation.

[0052] (3) By using the optimal path L(θ) to jointly optimize the path curvature, smoothness and energy consumption, the system can achieve stable operation in complex environments such as hospital corridors, elevators, corners and wards, and the path adjustment is smooth and does not disturb others. Attached Figure Description

[0053] Figure 1 A scene diagram illustrating an automatic following medical infusion method provided by the present invention;

[0054] Figure 2 This invention provides a schematic diagram of the structure of an automatic following medical infusion system;

[0055] Figure 3 A flowchart of an automatic following medical infusion method provided by the present invention;

[0056] Figure 4 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0057] Figure 5 This is a schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. Detailed Implementation

[0058] 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, and 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.

[0059] Please see Figure 1 , Figure 1 This is a scene diagram illustrating an automatic following medical infusion method provided by the intelligent sensor of the present invention. (Example) Figure 1 As shown, the terminal and server are connected via a network, such as a wired or wireless network. The terminal can include, but is not limited to, portable devices such as mobile phones and tablets with various network platform applications installed, as well as fixed terminals such as computers, kiosks, and advertising machines. The server provides users with various business services, including service push servers and user recommendation servers.

[0060] It should be noted that, Figure 1 The scenario diagram of an automatic follow-up medical infusion method shown is merely an example. The terminal, server, and application scenario described in the embodiments of the present invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of the present invention, and do not constitute a limitation on the technical solutions provided by the embodiments of the present invention. As those skilled in the art will know, with the evolution of the system and the emergence of new business scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0061] The terminal can be used for:

[0062] Based on the matching results between the RFID tag signal worn by the patient and the identity database, the binding strength between the patient and the infusion device is dynamically updated.

[0063] Based on the binding strength and the distance and speed data collected in real time by multiple sensors, the tracking accuracy of the patient calibrated by the binding strength is obtained;

[0064] Based on the tracking accuracy and the spatial topology data obtained by the environmental perception module, the optimal path for the movement of the infusion device is obtained;

[0065] Based on the feedback value of the optimal path and the tracking accuracy, the current speed of the adaptively controlled infusion device is obtained;

[0066] Based on the current speed and obstacle scanning data from the three-dimensional lidar, the threat level of the obstacle is obtained, and the obstacle avoidance decision value of the infusion device is determined based on the threat.

[0067] The motion control quantity of the infusion device is obtained based on the collaborative feedback of the optimal path, current speed, and obstacle avoidance decision value of the infusion device.

[0068] In addition, to achieve the above objectives, the present invention also proposes an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby realizing an automatic follow-up medical infusion method as described above.

[0069] Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of an automatic following medical infusion system provided by the present invention.

[0070] like Figure 2 As shown in the figure, an automatic following medical infusion system proposed in this embodiment of the invention includes:

[0071] The binding strength module 201 is used to obtain the dynamically updated binding strength between the patient and the infusion device based on the matching result of the RFID tag signal worn by the patient and the identity database.

[0072] The tracking accuracy module 202 is used to obtain the tracking accuracy of the patient calibrated by the binding strength based on the binding strength and the distance and speed data collected in real time by multiple sensors.

[0073] The path planning module 203 is used to obtain the optimal path for the movement of the infusion device based on the tracking accuracy and the spatial topology data obtained by the environmental perception module.

[0074] The speed control module 204 is used to obtain the current speed of the infusion device under adaptive control based on the feedback value of the optimal path and the tracking accuracy.

[0075] The obstacle avoidance decision module 205 is used to obtain the threat level of the obstacle based on the current speed and the obstacle scanning data of the three-dimensional lidar, and to determine the obstacle avoidance decision value of the infusion device based on the threat.

[0076] The automatic following module 206 is used to obtain the motion control quantity of the infusion device based on the collaborative feedback of the optimal path, current speed and obstacle avoidance decision value of the infusion device.

[0077] In some embodiments, the binding strength module 201 can also be used for:

[0078] Obtain an initial binding coefficient that represents the initial connection strength calibration value between the patient and the infusion device;

[0079] Construct a time decay term based on the time decay factor and the current time.

[0080] Obtain the periodic fluctuation factor used to simulate the periodic fluctuations of the bound signal;

[0081] The initial binding coefficient is processed based on the integral value of the binding signal strength at historical moments, the time decay term, and the periodic fluctuation factor to obtain the binding strength.

[0082] The binding strength can be expressed as:

[0083] ,

[0084] in, It is the binding strength. It is the initial binding coefficient. It is the time decay factor. It is the periodic fluctuation coefficient. It is the binding frequency. It is a signal strength function. It refers to the current moment.

[0085] In the specific implementation This is the initial binding coefficient, a calibrated value for the initial connection strength between the system and the patient. The larger the value, the stronger the initial state. It is a time decay term, bound to decay over time (such as when the patient leaves or the signal is lost). The larger the value, the faster the decay. It is a periodic fluctuation factor that simulates the periodic fluctuations of the bound signal (such as antenna interference caused by patient movement). It is the cumulative signal strength, representing the total accumulated signal energy / strength from 0 to the current time.

[0086] when When constant (e.g., when the signal is stable), the integral term becomes This facilitates numerical calculations.

[0087] The following numerical examples illustrate this, assuming:

[0088] 2. Dimensionless;

[0089] =0.1, the unit is ;

[0090] =0.3, dimensionless;

[0091] =π / 5, the unit is rad / s;

[0092] 1.5, a constant;

[0093] t=10, unit: s.

[0094] The calculation process is as follows:

[0095] ≈0.3679;

[0096] 1 0.3 1 = 0.7;

[0097] 1.5 10 = 15;

[0098] 2.0 0.3679 0.7 15 = 2.0 0.25753 15 ≈ 2.0 3.8629≈7.7258.

[0099] In this invention, if If the signal is a time-varying function (such as a wave signal), the integral term becomes more complex and requires numerical or symbolic integration. If the signal is momentarily interrupted, =0 will directly result in B(t)=0 (representing binding failure).

[0100] In some embodiments, the tracking accuracy module 202 can also be used for:

[0101] Dynamic weights are configured for different sensors to obtain the sensor weights for each sensor.

[0102] Based on the sensor weights of each sensor, the relative distance and relative speed measured by each sensor are fused together;

[0103] The reliability of the fused data is adjusted based on the reliability index of each sensor, and the final accuracy is calibrated by combining the real-time binding strength to obtain the tracking accuracy.

[0104] The tracking accuracy can be expressed as:

[0105] ,

[0106] in, It's about tracking accuracy. It is the sensor weight of the i-th sensor. It is the distance influence factor of the i-th sensor. It is the normalized velocity influence factor of the i-th sensor. It is the reliability index of the i-th sensor. It is a normalized relative distance. It is the normalized relative velocity. It refers to the binding strength.

[0107] In the specific implementation It represents tracking accuracy; the higher the accuracy, the greater the system's "effort," indicating a larger difficulty or deviation in current positioning. It is the weight of the i-th sensor, representing the proportion of the sensor's influence on tracking accuracy. It is the normalized distance influence factor; the larger the value, the more sensitive the system is to distance errors. It is the speed influence factor; the larger the value, the more sensitive the system is to speed errors. It is the sensor reliability index, which represents the confidence level of the sensor's output results. It amplifies or suppresses error terms in an exponential form. It is the normalized relative distance, defined as: ,in The maximum distance measurement range is such that the actual measured relative distance between the target (patient) and the system (infusion equipment) is 0.5m. It is the normalized relative velocity, defined as: ,in The maximum measurable relative velocity is 0.2 m / s for the target (patient) relative to the system (infusion equipment). It is the binding strength, which comes from the previous step and represents the basic trustworthiness of the tracking. The larger the value, the more stable the system.

[0108] Now assume: maximum distance Maximum speed ,

[0109] Current measurements: ,

[0110] Conclusion: ,

[0111] Assume the parameter values ​​of the first sensor are as follows:

[0112] , , , ,

[0113] Bond strength ;

[0114] Then the first sensor pair The contributions are as follows:

[0115] ,

[0116] It's understandable that if there are multiple sensors, the data from the other sensors are summed up sequentially to obtain the multi-sensor result. The value is sufficient; further details will not be provided here.

[0117] In the examples of this invention, due to the binding strength The value is relatively high, and the denominator is large, so the system considers the current target tracking "reliable" and does not require much correction intervention. If the patient identity binding weakens (e.g.) (decline), then the same deviation will lead to a higher This means that the "tracking error is amplified," triggering more aggressive control behaviors (such as accelerating to approach, path correction, etc.).

[0118] In some embodiments, the path planning module is further configured to:

[0119] Based on the path curvature function and tracking accuracy of each path, a control overhead term for positioning and tracking accuracy is constructed;

[0120] Calculate the square of the rate of change of the angle of the infusion device traveling along each of the aforementioned paths;

[0121] Calculate the unit energy consumption of the infusion device under each of the path controls;

[0122] The optimal path is obtained by combining the control overhead, the square of the angle change rate, and the unit energy consumption, and then performing differentiation.

[0123] The optimal path can be represented as:

[0124] ,

[0125] in, It is the optimal path. It is the normalized path curvature. It is the smoothness weighting coefficient. It is the energy consumption weighting coefficient. It is normalized energy consumption. It is the planning cycle. It is the normalized tracking accuracy.

[0126] In the specific implementation It represents the optimal path, the goal of path planning optimization; the smaller the value, the better the path. This is the normalized tracking accuracy, representing the current accuracy of target tracking. The greater the "uncertainty", the larger this value becomes. The normalized path curvature is calculated using the path curvature function. The more convoluted the path, the greater the curvature, which may affect feasibility and stability. It is the square of the rate of change of the angle of the infusion equipment traveling along each path, weighted smoothness, avoiding "sharp turns" in the path, and enhancing the patient experience and control stability. It is the normalized unit energy consumption under path control, including wheel set control energy consumption, acceleration loss, etc. It is the planning cycle, which is generally the local path planning update cycle (e.g., 2–5 seconds).

[0127] It is the control overhead item for positioning and tracking accuracy, when the target motion is uncertain ( (For large areas), the path needs to avoid high curvature regions to reduce the probability of loss. It improves path smoothness, reduces sharp turns, and enhances the stability of infusion equipment. Used to minimize energy consumption, reduce operating costs within hospitals, and improve battery life.

[0128] It can be understood that the path planning in this invention is essentially about finding a function. , making Minimum. This type of problem can be solved using variational methods or shortest path optimization algorithms (such as dynamic programming, reinforcement learning, etc.).

[0129] Assuming the system is planned along an approximately constant curvature path (such as a uniform turning speed), the integral function can be simplified and calculated:

[0130] ;

[0131] ;

[0132] ;

[0133] ;

[0134] ;

[0135] Then the integrand within the integral is:

[0136] ;

[0137] Assuming T = 10s, then:

[0138] This indicates that, under the current tracking accuracy, path curvature, velocity smoothness, and energy consumption weighting conditions, the total cost of this path is 8.55.

[0139] In summary, in this invention, if a larger curvature (sharp turn) is used, that is... Increasing this cost increases the first factor, which is detrimental to accurate tracking. To reduce energy consumption (by slowing down while moving straight), you can control... Reduce and optimize the overall path. Parameters μ and γ can be configured according to the hospital's actual preferences, such as emphasizing stability and energy saving.

[0140] In some embodiments, the speed control module 204 can also be used for:

[0141] The tracking accuracy is adjusted based on the velocity response coefficient to obtain the adjusted tracking accuracy.

[0142] Calculate the rate of change of the binding strength over time;

[0143] Calculate the cumulative path cost based on the path impact factor and the optimal path at each historical moment;

[0144] The current speed is determined based on the adjusted tracking accuracy, the time rate of change of the binding strength, and the cumulative path cost.

[0145] The current speed of the infusion device can be expressed as:

[0146] ,

[0147] in, This is the current speed. It is the velocity response coefficient. It is a path influence factor. It is the normalized tracking accuracy. It is the normalized average of the path quality. , or can be written as , It is the normalized cumulative path cost. It is the normalized rate of change of binding strength, that is, the rate of change of binding strength over time. ,in It is the maximum rate of change.

[0148] In practice, this formula is the core function in the control system used to adjust the following speed in real time, combining dynamic information such as binding strength changes, target tracking accuracy, and path planning.

[0149] This reflects the system's immediate speed response based on the current tracking accuracy and the changing trend of the patient's binding status. When binding is enhanced and tracking accuracy is high, the system tends to increase speed to closely follow the patient; conversely, if binding status declines or tracking becomes unreliable, this value approaches 0 or a negative value, reducing speed to ensure safety and accuracy.

[0150] This represents the system's consideration of the overall feasibility and quality of the path: when the path score is high (i.e., the path is smooth, safe, and energy-efficient), the system can choose to increase the speed; if the path is complex, has a large curvature, or has high energy consumption, the speed will automatically decrease to ensure smooth travel and energy efficiency and safety.

[0151] Now assume the values ​​of each parameter are as follows:

[0152] ;

[0153] ;

[0154] ;

[0155] ;

[0156] but:

[0157] .

[0158] This indicates the current speed is 1.44 m / s, which is a typical indoor walking speed (suitable for a hospital environment). When the binding status deteriorates... When the speed is close to 0 or the path cost is too high, the speed will naturally decrease → improve the robustness of the system. All parameters are adjustable to adapt to the needs of different departments, patient groups or hospital environments.

[0159] In summary, the speed control of this invention is not simply a proportional adjustment based on the target distance or current position, but rather a joint control through two main lines: dynamic binding change and path cost accumulation. This achieves rapid response to binding status (such as device start-up, re-capture after losing track, etc.), smooth operation path (avoiding unnecessary rapid movement due to tortuous or complex environments), and reliable tracking by combining the accuracy of multiple sensors.

[0160] In some embodiments, the obstacle avoidance decision module 205 can also be used for:

[0161] Obtain the distance normalized scale and the azimuth normalized scale;

[0162] Based on the distance normalization scale, the distance between the infusion device and each obstacle, and the reference distance, a distance influence factor is constructed.

[0163] Based on the azimuth normalization scale, the azimuth of each obstacle, and the azimuth of the infusion device, an angle influence factor is constructed.

[0164] The obstacle avoidance decision value is determined based on the distance influence factor and the angle influence factor.

[0165] The obstacle avoidance decision value can be expressed as:

[0166] ,

[0167] ,

[0168] in, It is the obstacle avoidance decision value. is the distance between the infusion device and the j-th obstacle, and r is the reference distance between the infusion device and the reference point (such as the origin). It is the current azimuth angle. It is the azimuth angle of the j-th obstacle. It is a distance normalization scale. It is an azimuth normalized scale. This is the current speed. is the normalized threat level of the j-th obstacle, and m is the total number of obstacles.

[0169] In the specific implementation This is the distance influence factor, which represents the relative radial distance error between the current device and the obstacle. This is the angle influence factor, which represents the relative angular deviation between the equipment and the obstacle. The larger the deviation, the more likely the obstacle is not directly in front or in a non-primary direction.

[0170] It is a two-dimensional Gaussian function centered on the obstacle. , Construct a decay field:

[0171] The closer the distance or angle is to the center of the obstacle, the larger the value, indicating that "it is more necessary to avoid it";

[0172] When the distance or angle deviates from the center of the obstacle, the value decays rapidly, indicating "no need for excessive avoidance".

[0173] Now assume:

[0174] Two obstacles;

[0175] ;

[0176] Obstacle 1: , , , , ;

[0177] Obstacle 2: , , Similarly, normalization is applied.

[0178] Assumption ,so ;

[0179] Current speed ,

[0180] calculate:

[0181] For obstacle 1:

[0182] ;

[0183] ;

[0184] For obstacle 2:

[0185] ;

[0186] ;

[0187] merge:

[0188] .

[0189] The current obstacle avoidance decision value is 1.506. A higher value indicates stronger obstacle avoidance interference ahead, and the system should immediately: reduce speed or stop abruptly, trigger path replanning, or issue a safety warning (such as voice or light prompts). Therefore, this invention allows for precise control over the impact of each obstacle and reflects the actual risk differences of different types of obstacles (people, walls, moving objects).

[0190] In some embodiments, the automatic following module 206 can also be used for:

[0191] Calculate the rate of change of the optimal path over time;

[0192] Based on the rate of change of the optimal path over time, the current speed, and the obstacle avoidance decision value, and in conjunction with the control gain coefficients corresponding to each of the above parameters, the motion control quantity is determined.

[0193] The motion control quantity can be expressed as:

[0194] ,

[0195] in, It is the amount of motor control. It is the normalized current velocity. It is a normalized obstacle avoidance decision value. It is the normalized rate of change of the optimal path over time. These are the first control gain coefficient, the second control gain coefficient, and the third control gain coefficient.

[0196] In the specific implementation It is a motion control quantity, which is ultimately output to the drive motor (wheel set) as a comprehensive control command to control the motion behavior of the infusion system. It is the normalized current velocity, the current expected velocity obtained based on the binding strength and path integral (reflecting the follower intent). It is a normalized obstacle avoidance decision value, which represents the risk level of the obstacle ahead; the larger the value, the more dangerous it is. It is the normalized rate of change of the optimal path over time, representing the rate of change of path complexity. Higher dynamics require a faster response. Determining the degree of influence of each subsystem on the final motion control requires experimental parameter tuning and optimization.

[0197] Assume the values ​​are as follows:

[0198] m / s, ;

[0199] , ;

[0200] , ;

[0201] Normalization results:

[0202] ;

[0203] ;

[0204] ;

[0205] final:

[0206] ,

[0207] Assumption If the values ​​are 0.5, 0.3, and 0.2 respectively, then:

[0208]

[0209] In summary, this invention uses this formula to weight and fuse three key factors—speed, obstacle avoidance, and path change—to construct a unified, adjustable, and consistent control output, effectively improving the system's dynamic response and stability in complex environments. This design achieves the optimal balance between following accuracy, safe obstacle avoidance, and smooth movement in the intelligent infusion system, ensuring the safety, comfort, and adaptability of the patient's movement. It is the core control foundation for realizing the automatic following function.

[0210] Please see Figure 3The present invention provides a flowchart of an automatic follow-up medical infusion method, comprising the following steps:

[0211] Step 301: Based on the matching results between the RFID tag signal worn by the patient and the identity database, obtain the dynamically updated binding strength between the patient and the infusion device;

[0212] Step 302: Based on the binding strength and the distance and speed data collected in real time by multiple sensors, obtain the tracking accuracy of the patient calibrated by the binding strength;

[0213] Step 303: Based on the tracking accuracy and the spatial topology data obtained by the environmental perception module, obtain the optimal path for the movement of the infusion device;

[0214] Step 304: Based on the feedback value of the optimal path and the tracking accuracy, obtain the current speed of the adaptively controlled infusion device;

[0215] Step 305: Based on the current speed and the obstacle scanning data from the three-dimensional lidar, obtain the threat level of the obstacle, and determine the obstacle avoidance decision value of the infusion device based on the threat level;

[0216] Step 306: Based on the collaborative feedback of the optimal path, current speed, and obstacle avoidance decision value of the infusion device, obtain the motion control quantity of the infusion device.

[0217] It should be noted that for the specific embodiments and beneficial effects of steps 301-305 above, please refer to the relevant descriptions of modules 201-205 above, which will not be repeated here.

[0218] Please see Figure 4 , Figure 4 This is a schematic diagram illustrating an embodiment of the electronic device provided in this invention. For example... Figure 4 As shown, an embodiment of the present invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, it performs the following steps:

[0219] Based on the matching results between the RFID tag signal worn by the patient and the identity database, the binding strength between the patient and the infusion device is dynamically updated.

[0220] Based on the binding strength and the distance and speed data collected in real time by multiple sensors, the tracking accuracy of the patient calibrated by the binding strength is obtained;

[0221] Based on the tracking accuracy and the spatial topology data obtained by the environmental perception module, the optimal path for the movement of the infusion device is obtained;

[0222] Based on the feedback value of the optimal path and the tracking accuracy, the current speed of the adaptively controlled infusion device is obtained;

[0223] Based on the current speed and obstacle scanning data from the three-dimensional lidar, the threat level of the obstacle is obtained, and the obstacle avoidance decision value of the infusion device is determined based on the threat.

[0224] The motion control quantity of the infusion device is obtained based on the collaborative feedback of the optimal path, current speed, and obstacle avoidance decision value of the infusion device.

[0225] In addition, to achieve the above objectives, the present invention also proposes an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby realizing an automatic follow-up medical infusion method as described above.

[0226] Please see Figure 5 , Figure 5 This is a schematic diagram illustrating an embodiment of a computer-readable storage medium provided by an embodiment of the present invention. For example... Figure 5 As shown, this embodiment provides a computer-readable storage medium 500 on which a computer program 411 is stored. When the computer program 411 is executed by a processor, it performs the following steps:

[0227] Based on the matching results between the RFID tag signal worn by the patient and the identity database, the binding strength between the patient and the infusion device is dynamically updated.

[0228] Based on the binding strength and the distance and speed data collected in real time by multiple sensors, the tracking accuracy of the patient calibrated by the binding strength is obtained;

[0229] Based on the tracking accuracy and the spatial topology data obtained by the environmental perception module, the optimal path for the movement of the infusion device is obtained;

[0230] Based on the feedback value of the optimal path and the tracking accuracy, the current speed of the adaptively controlled infusion device is obtained;

[0231] Based on the current speed and obstacle scanning data from the three-dimensional lidar, the threat level of the obstacle is obtained, and the obstacle avoidance decision value of the infusion device is determined based on the threat.

[0232] The motion control quantity of the infusion device is obtained based on the collaborative feedback of the optimal path, current speed, and obstacle avoidance decision value of the infusion device.

[0233] In addition, to achieve the above objectives, the present invention also proposes an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby realizing an automatic follow-up medical infusion method as described above.

[0234] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0235] Those skilled in the art will understand that embodiments of the present invention can be provided as systems, methods, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0236] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0237] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction set implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0238] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0239] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0240] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An automatic following medical infusion system, characterized in that, The system includes: The binding strength module is used to obtain a dynamically updated binding strength between the patient and the infusion device based on the matching result of the RFID tag signal worn by the patient and the identity database. It is also used to obtain an initial binding coefficient representing the initial connection strength calibration value between the patient and the infusion device; construct a time decay term based on a time decay factor and the current time; obtain a periodic fluctuation factor to simulate the periodic fluctuation of the binding signal; and process the initial binding coefficient based on the integral value of the binding signal strength at historical times, the time decay term, and the periodic fluctuation factor to obtain the binding strength. The tracking accuracy module is used to obtain the tracking accuracy of the patient calibrated by the binding strength based on the binding strength and the distance and speed data collected in real time by multiple sensors. The path planning module is used to obtain the optimal path for the infusion device to move based on the tracking accuracy and the spatial topology data obtained by the environmental perception module. The speed control module is used to obtain the current speed of the adaptively controlled infusion device based on the feedback value of the optimal path and the tracking accuracy; it is also used to adjust the tracking accuracy based on the speed response coefficient to obtain the adjusted tracking accuracy; calculate the time change rate of the binding strength; calculate the cumulative path cost based on the path influence factor and the optimal path at each historical moment; and determine the current speed based on the adjusted tracking accuracy, the time change rate of the binding strength, and the cumulative path cost. The obstacle avoidance decision module is used to obtain a distance normalized scale and an azimuth normalized scale. Based on the distance normalized scale, the distance between the infusion device and each obstacle, and a reference distance, a distance influence factor is constructed. Based on the azimuth normalized scale, the azimuth of each obstacle, and the azimuth of the infusion device, an angle influence factor is constructed. Based on the distance influence factor and the angle influence factor, an obstacle avoidance decision value is determined, wherein the obstacle avoidance decision value is expressed as: in, It is the obstacle avoidance decision value. is the distance between the infusion device and the j-th obstacle, and r is the reference distance between the infusion device and the reference point. It is the current azimuth angle. It is the azimuth angle of the j-th obstacle. It is a distance normalization scale. It is an azimuth normalized scale. This is the current speed. is the normalized threat level of the j-th obstacle, and m is the total number of obstacles; An automatic following module is used to obtain the motion control quantity of the infusion device based on the coordinated feedback of the optimal path, current speed and obstacle avoidance decision value of the infusion device.

2. The automatic following medical infusion system according to claim 1, characterized in that, The tracking accuracy module is also used for: Dynamic weights are configured for different sensors to obtain the sensor weights for each sensor. Based on the sensor weights of each sensor, the relative distance and relative speed measured by each sensor are fused together; The reliability of the fused data is adjusted based on the reliability index of each sensor, and the final accuracy is calibrated by combining the real-time binding strength to obtain the tracking accuracy.

3. The automatic following medical infusion system according to claim 1, characterized in that, The path planning module is also used for: Based on the path curvature function and tracking accuracy of each path, a control overhead term for positioning and tracking accuracy is constructed; Calculate the square of the rate of change of the angle of the infusion device traveling along each of the aforementioned paths; Calculate the unit energy consumption of the infusion device under each of the path controls; The optimal path is obtained by combining the control overhead, the square of the angle change rate, and the unit energy consumption, and then performing differentiation.

4. The automatic following medical infusion system according to claim 1, characterized in that, The automatic following module is also used for: Calculate the rate of change of the optimal path over time; The motion control quantity is determined based on the rate of change of the optimal path over time, the current speed, and the obstacle avoidance decision value.

5. An automatic following-type medical infusion method, wherein the method implements the system as described in claim 1, characterized in that, The method includes: Based on the matching results between the RFID tag signal worn by the patient and the identity database, the dynamically updated binding strength between the patient and the infusion device is obtained, and an initial binding coefficient is acquired to represent the initial connection strength calibration value between the patient and the infusion device. A time decay term is constructed based on the time decay factor and the current time. A periodic fluctuation factor is acquired to simulate the periodic fluctuation of the binding signal. The initial binding coefficient is processed based on the integral value of the signal strength of the binding signal at historical times, the time decay term, and the periodic fluctuation factor to obtain the binding strength. Based on the binding strength and the distance and speed data collected in real time by multiple sensors, the tracking accuracy of the patient calibrated by the binding strength is obtained; Based on the tracking accuracy and the spatial topology data obtained by the environmental perception module, the optimal path for the movement of the infusion device is obtained; Based on the feedback values ​​of the optimal path and the tracking accuracy, the current speed of the adaptively controlled infusion device is obtained; the tracking accuracy is adjusted according to the speed response coefficient to obtain the adjusted tracking accuracy; the time change rate of the binding strength is calculated; the cumulative path cost is calculated based on the path influence factor and the optimal path at each historical moment; and the current speed is determined based on the adjusted tracking accuracy, the time change rate of the binding strength, and the cumulative path cost. Obtain the distance normalized scale and azimuth normalized scale. Based on the distance normalized scale, the distance between the infusion device and each obstacle, and the reference distance, construct a distance influence factor. Based on the azimuth normalized scale, the azimuth of each obstacle, and the azimuth of the infusion device, construct an angle influence factor. Based on the distance influence factor and the angle influence factor, determine the obstacle avoidance decision value, whereby the obstacle avoidance decision value is expressed as: in, It is the obstacle avoidance decision value. is the distance between the infusion device and the j-th obstacle, and r is the reference distance between the infusion device and the reference point. It is the current azimuth angle. It is the azimuth angle of the j-th obstacle. It is a distance normalization scale. It is an azimuth normalized scale. This is the current speed. is the normalized threat level of the j-th obstacle, and m is the total number of obstacles; The motion control quantity of the infusion device is obtained based on the collaborative feedback of the optimal path, current speed, and obstacle avoidance decision value of the infusion device.

Images