Simulation machine wounded person system based on search and rescue environment perception

Abstract

The invention, which relates to the technical field of robot motion control, discloses a robot wounded person simulation system based on search and rescue environment perception, comprising an environment acquisition module for acquiring environment information in a water environment area where a robot wounded person is located, the physiological mapping module is used for predicting the physiological evolution trend of the machine wounded person in the current environment based on the environment state vector, the injury condition prediction module is used for predicting the evolution trend of physiological and injury condition states in a future time period through a built injury condition evolution model, and the operation control module is used for executing corresponding physiological actions. A physiological driving model and an environment injury condition coupling mechanism are utilized, a machine wounded person can autonomously adjust floating postures, limb actions, respiratory rhythms and the like according to environment changes and time evolution, meanwhile, multiple environment factors are considered, injury condition development is predicted through the physiological environment coupling model, and the system is suitable for being popularized and applied. The comprehensive simulation of the physiological state, the injury state and the environmental condition is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot action control, and particularly relates to a simulation machine wounded personnel system based on search and rescue environment sensing. BACKGROUND

[0002] In current aviation search and rescue simulation training, the simulation of the distressed personnel mainly depends on fixed injury mannequins or simple mechanical devices. Such devices can only show the external physical characteristics of the wounded personnel, but lack the dynamic evolution ability of vital signs, and cannot truly simulate the physiological changes and injury development of the distressed personnel in the sea environment.

[0003] In the sea environment, factors such as seawater temperature, oxygen concentration and hydrodynamic force will jointly act on the fallen personnel, significantly affect the body temperature regulation, respiratory function and overall physiological state of the fallen personnel, and further change the evolution trend of the injury.

[0004] More importantly, the existing simulation technology cannot accurately reproduce the process of the body temperature change of the distressed personnel with time, which directly leads to the fact that in search and rescue drills, the infrared detection equipment cannot be effectively coordinated, and the reality of the training and the effect of the equipment linkage are weakened. Especially at night or in low-visibility conditions, infrared detection is the main means of search and rescue, and the mannequin lacking response to body temperature changes cannot meet this training requirement. SUMMARY

[0005] I) Technical problems solved The present application provides a simulation machine wounded personnel system based on search and rescue environment sensing, which can make the physiological response and injury of the machine wounded personnel continuously evolve with time, and correspondingly drive the related actions of the machine wounded personnel.

[0006] II) Technical solutions To achieve the above-mentioned purpose, the present application provides the following technical solutions: a simulation machine wounded personnel system based on search and rescue environment sensing, comprising: An environment acquisition module is configured to acquire environmental information in a water environment region where the machine wounded personnel is located, including dissolved oxygen concentration, hydrodynamic disturbance intensity, and temperature data captured by a plurality of temperature sensing detection components distributed on the machine wounded personnel, and to construct a corresponding environmental state vector after feature extraction of the environmental information at each time step. The physiological mapping module receives the environmental state vector in real time, combines a heat balance model and an oxygen metabolism model of the human body, calculates the instantaneous physiological response of the machine wounded person in the current environment, and generates a simulation state vector of the current time step; wherein, according to the temperature mean value and gradient in the temperature data, the instantaneous body temperature change is calculated by using the heat balance model; according to the dissolved oxygen concentration and the water dynamic disturbance index, the ventilation rate and the anoxia risk index are calculated by combining the oxygen metabolism model; The injury evolution prediction module takes the environmental state vector and the simulation state vector of the historical and current time steps as inputs, calculates the physiological change rates including body temperature, ventilation rate, anoxia risk and energy reserve based on the heat balance model, the oxygen metabolism model and the energy consumption dynamics model, and predicts the injury evolution trend combined with the pathological mechanism model; the physiological change rates and the injury evolution trend are superimposed on the simulation state vector of the current time step to predict the simulation state vector of the next time step; The operation control module maps the predicted simulation state vector of the next time step into the physiological performance parameters of the machine wounded person, and converts it into the action execution control signal corresponding to the next time step.

[0007] Further, the environment acquisition module collects the external environmental conditions of the water environment and the surface temperature analysis of itself in real time through the sensing units distributed in the machine wounded person and the outside, specifically including: The dissolved oxygen concentration of the current time step in the water body is obtained through the dissolved oxygen sensor; The local temperature is captured through the temperature sensing detection components arranged on the head, trunk and limbs of the machine wounded person respectively to form a local temperature sequence, and the temperature gradient and the average temperature are calculated through the temperature difference of each temperature sensing detection component area; The water flow acceleration around the machine wounded person is detected through the inertial measurement unit as the water dynamic disturbance intensity.

[0008] Further, the environmental state vector constructed includes four dimensional variables of average water temperature, temperature gradient, dissolved oxygen concentration and water dynamic disturbance intensity, and each dimension of the variable is standardized through a feature extraction function to form a unified vector form: ; Among them, is the average water temperature of the current time step , is the temperature gradient of the current time step , is the dissolved oxygen concentration of the current time step , is the water dynamic disturbance intensity of the current time step .

[0009] Further, the physiological mapping module obtains the environmental state vector output by the environment acquisition module, combines a human body heat balance model and an oxygen metabolism model, predicts a physiological evolution trend of the machine wounded person in the current environment, and generates the simulation state vector; specifically: According to the average water temperature and temperature gradient of the water environment, combined with the heat balance model, the change rate of the body temperature of the set core area of the machine wounded person is calculated; According to the oxygen concentration and hydrodynamic disturbance intensity of the water environment, combined with the oxygen metabolism model and human body respiratory dynamics, the ventilation rate and hypoxia risk index of the current time step are calculated; Based on the shivering consumption and limb activity intensity of the human body, the residual energy change rate of the machine wounded person at the current time step is calculated.

[0010] Further, the simulation state vector constructed includes four-dimensional variables such as body temperature, ventilation rate, hypoxia index, and energy reserve, and the form of the simulation state vector is: ; Among them, the body temperature of the set core area And the energy reserve , respectively, through the change rate of the core body temperature and the residual energy change rate of the current time step , superimposed on the corresponding dimensional variables of the simulation state vector of the next time step , to predict the corresponding dimensional variables of the next time step ; The hypoxia index converted from the hypoxia risk index of the current time step , The ventilation rate of the current time step .

[0011] Further, the injury evolution prediction module receives the environmental state vector and the simulation state vector of the current time step and a set number of historical time steps, and performs standardization processing on the input vectors, and calculates the injury evolution trend by using a pathological model; specifically including: According to the heat balance model, the deviation of the core body temperature from the standard physiological value is calculated, and the low-temperature stress pathological model is used to convert the deviation into a tissue function decline rate; Combined with the ventilation rate, the hypoxia risk index, and the oxygen supply and demand balance mechanism, the hypoxia degree is calculated, and the pathological mechanism is mapped to the injury probability; Combined with the residual energy, activity intensity and shivering consumption, the energy depletion rate is calculated, and the energy-related injury evolution rate is mapped by the pathological model.

[0012] Further, the physiological change rate is superimposed on the simulation state vector of the current time step: ; Meanwhile, the injury evolution trend is mapped as an injury state component: ; The simulation state vector of the next time step is formed comprehensively: .

[0013] Further, the operation control module maps the simulation state vector of each time step into high-level action parameters, and the high-level action parameters include adjusting body temperature, floating posture, breathing rhythm, and adjusting limb movement amplitude.

[0014] Further, the high-level action parameters are converted into low-level execution instructions, and the low-level action instructions correspond to the control amount of specific execution components; specifically including: controlling the heating and heat dissipation power of the temperature sensing simulator arranged in the machine wounded person, corresponding to the change of body temperature; adjusting the valve opening of the air bag arranged in the machine wounded person to control the buoyancy and floating height; controlling the angle and speed of the driving motor arranged in the limbs of the machine wounded person respectively, corresponding to the limb movement amplitude; controlling the output of the breathing simulator arranged in the machine wounded person, corresponding to the ventilation rhythm.

[0015] III) Beneficial effects: Compared with the prior art, the present application has the following beneficial effects: The present application can continuously advance the physiological response and injury evolution of the machine wounded person with time, rather than staying in a static state, and can map this dynamic change into action control signals, so that the simulation result of the machine wounded person in the sea rescue environment is closer to the real physiological change.

[0016] By receiving the environment state vector and the simulation state vector at each time step, the physiological change rate is calculated based on the heat balance, oxygen metabolism and energy consumption dynamics model, and the injury evolution trend is predicted combined with the pathological mechanism model. The simulation state is dynamically updated through the calculation result, so that the physiological and injury performance of the machine wounded person not only embodies the continuous evolution of the time dimension, but also reflects the real-time influence of environmental disturbance.

[0017] In addition, the operation control module converts the predicted simulation state of the next time step into corresponding physiological performance parameters, and adjusts the floating posture, limb movement, breathing rhythm, etc. of the machine wounded person after mapping into action control signals, showing the self-adaptive simulation characteristics of the machine wounded person to the search and rescue environment. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 A principle block diagram of a simulated machine wounded person system based on search and rescue environment sensing provided by the embodiment of the present application; Figure 2 A flowchart of a simulation machine wounded personnel system based on search and rescue environment perception provided by an embodiment of the present application is shown in the figure. Figure 3 In a simulation machine wounded personnel system based on search and rescue environment perception provided by an embodiment of the present application, a principle block diagram of the operation control module is shown in the figure. Figure 4 In a simulation machine wounded personnel system based on search and rescue environment perception provided by an embodiment of the present application, a flowchart from the acquisition of environmental data and physiological state data to the execution of corresponding actions of the machine wounded personnel is shown in the figure. In the figure: 100, environment acquisition module; 200, physiological mapping module; 300, wound evolution prediction module; 400, operation control module. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0020] In the description of the present application, it should be understood that the terms "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0021] In addition, if the terms "first", "second", etc. are used only for differentiation and cannot be understood as indicating or implying relative importance.

[0022] It should be noted that the features in the embodiments of the present application can be combined with each other without conflict.

[0023] In aviation search and rescue training and simulated scenarios of distressed personnel, the present inventor found that the search and rescue task has a high degree of uncertainty, and the physiological state and wound condition of the distressed personnel often change significantly over time. The existing simulation system lacks dynamic wound condition prediction and simulation capabilities based on time dimension and environmental conditions, and cannot drive the wound evolution process at different time steps. This deficiency limits the restoration of complex situations in search and rescue training and reduces the verification value of the effectiveness of rescue strategies.

[0024] In combination Figures 1 to 4 The system shown in the simulation machine wounded system based on search and rescue environment perception, through dynamic physiological state and injury evolution simulation, the simulation of injury can face closer to the real distress situation, improve the scientific nature and effectiveness of search and rescue decision, provide quantifiable, available dynamic data basis for search and rescue strategy optimization and training efficiency evaluation.

[0025] Specifically, first, the environment acquisition module 100 for acquiring environmental data, in combination Figure 1 And Figure 2 As shown, the module is used to acquire the key environmental information in the water environment area where the machine wounded is located, and to extract the features of each type of information to form an environmental state vector which can be used for subsequent physiological mapping and injury evolution prediction.

[0026] In the embodiment of the application, the environmental state vector includes four dimensional variables: average water temperature, temperature gradient, dissolved oxygen concentration and water dynamic disturbance intensity.

[0027] The core goal of the environment acquisition module 100 is to realize the accurate perception of the external water environment condition and the local state of the machine wounded, and to provide input basis for the adaptive injury evolution of the machine wounded.

[0028] More specifically, in some embodiments of the application, the environment acquisition module 100 measures the oxygen content in the water body in real time through the dissolved oxygen sensor arranged near the neck of the machine wounded or near the mouth and nose, and the dissolved oxygen sensor outputs the oxygen concentration data (unit ) in the water body, which can be used to evaluate the potential hypoxia risk. For example, in the sea floating training, if the sensor measures that the dissolved oxygen concentration in the water body is lower than 5 , it indicates that the machine wounded may enter the hypoxia state, and the hypoxia index will be adjusted in the subsequent simulation state.

[0029] Therefore, it can be understood that the collection of dissolved oxygen by the environment acquisition module 100 can perceive the change of water oxygenation in real time, so as to capture the immediate influence of environmental mutation on the injury, and provide dynamic input for the physiological mapping module 200.

[0030] Regarding the collection of seawater temperature by the environment acquisition module 100 and the analysis of temperature gradient, specifically, in some embodiments of the application, temperature sensing detection components are installed on the head, trunk and limbs of the machine wounded to form a local temperature collection network. In this way, the data collected from each temperature sensing point forms a local temperature sequence, and the temperature gradient is obtained by calculating the temperature difference of each point, and the average temperature is calculated to form a temperature feature vector.

[0031] For example, if at a certain moment the temperature of a machine-damaged person's head is 30°C, torso is 28°C, and limbs are 26°C, the average temperature can be calculated to be 28°C, with a temperature gradient of 4°C. This temperature gradient information can be understood as being used in thermal equilibrium models to determine the degree of cold exposure in different parts of the body, influencing core body temperature prediction and shiver intensity estimation.

[0032] In summary, by analyzing local temperature distribution and gradients, we can more accurately simulate the dynamics of human body thermal balance, and reflect the overall thermal stress state better than single-point temperature measurement.

[0033] Regarding the environmental acquisition module 100's acquisition of hydrodynamic disturbance intensity, specifically, it uses an inertial measurement unit (IMU) or accelerometer inside the machine to detect the acceleration and fluctuation characteristics of the water flow around the machine in real time. The measured acceleration vector is then converted into a hydrodynamic disturbance index to describe the external effects of the surrounding water on the machine's buoyancy and attitude.

[0034] To continue with the example, suppose there is a water flow acceleration of 0.5 m / s² on the sea surface. This indicates that the water flow is significantly disturbing the floating of the robot's injured crew, requiring adjustments to the floating posture and limb movement amplitude in the subsequent motion control module. Therefore, it is understandable that by collecting data on the intensity of hydrodynamic disturbances, dynamic water environment disturbance data is provided, enabling the robot's injured crew to adapt in real time during floating and limb movements, thus improving the realism of the injury evolution simulation.

[0035] After feature extraction from the above three types of collected information, an environmental state vector is formed: ; in, For the current moment Average water temperature For the current moment The temperature gradient is taken into account that the human body's temperature regulation is directly related to the temperature difference of the water body. The calculation of these two indicators comes from the heat conduction equation and the theory of convection heat transfer. Here, it is necessary to first set the temperature of the seawater as the amount of heat conduction exchange.

[0036] For the current moment Dissolved oxygen concentration is directly coupled with hypoxia indicators, and its calculation is based on the physiological functional relationship between oxygen partial pressure and blood oxygen saturation.

[0037] For the current moment The intensity of hydrodynamic disturbance can be established using an energy consumption model in fluid mechanics and experimental regression data to create an empirical formula.

[0038] The environmental state vector is taken as an input of a subsequent physiological mapping module 200 to realize mapping of environmental information to a simulation state.

[0039] In summary, it can be understood that the environmental acquisition module 100 fuses temperature, oxygen concentration, and hydrodynamic disturbance to form a comprehensive feature vector, which more comprehensively reflects environmental conditions, and the environmental state vector can be directly taken as an input of a subsequent physiological mapping and injury evolution module to realize data closed loop and iterative prediction.

[0040] Regarding the physiological mapping module 200, a core function of the module is to map the environmental state vector output by the environmental acquisition module 100 to a simulation state vector of the machine-injured person, and predict the physiological evolution trend under the current environment.

[0041] In some embodiments of the present application, the simulation state vector contains four dimensional variables: core body temperature, ventilation rate, hypoxia index, and energy reserve.

[0042] The physiological mapping module 200 realizes dynamic mapping of environmental information to physiological state by combining a human body heat balance model, an oxygen metabolism model, and an energy consumption model, and provides direct input for injury evolution prediction.

[0043] Specifically, first regarding the core body temperature change rate calculation, the average water temperature and temperature gradient in the environmental state vector are needed to be obtained, and the human body heat balance model, that is, the above-mentioned factors such as body surface heat exchange, seawater heat conduction convection, etc. are used to calculate the instantaneous change rate of core body temperature. In some embodiments of the present application, the specific calculation formula is as follows:

[0044] ; wherein, is the heat capacity of the human body, is the convective heat transfer coefficient, is the body surface area, which are all known parameters; is the current core body temperature, represents the local heat flow caused by the temperature gradient, is the metabolic heat production, is the energy consumed by movement, which can be obtained by corresponding conventional calculation, and is not specifically limited here.

[0045] Regarding the ventilation rate and hypoxia risk prediction, the dissolved oxygen concentration and hydrodynamic disturbance intensity in the environmental state vector are first obtained, and then the oxygen metabolism model and respiratory dynamics are combined to predict the current ventilation rate and hypoxia risk index. In some embodiments of the present application, the specific calculation formula is as follows:

[0046] ; ; wherein, is the ventilation rate at the resting moment, generally selected as the average ventilation rate when the heart rate is stable in the normal range, is the reference oxygen concentration, is the function of mapping the hypoxia risk, and is the environmental response coefficient, and the data acquisition includes but is not limited to the physiological experimental data, and then converted into a mathematical expression, which is not limited to specific numerical values here.

[0047] For example, it is assumed that the dissolved oxygen concentration in water is 4 , the hydrodynamic disturbance intensity is 0.5 , the ventilation rate is increased by 20% combined with the respiratory kinetics model, the hypoxia risk index is increased to 0.65 (scale 0-1), and an index is provided for subsequent injury evolution prediction.

[0048] Regarding the energy reserve change rate calculation, it is specifically based on the human energy consumption kinetics model to calculate the current energy reserve change rate. The calculation formula is a general formula, which is not limited here.

[0049] It is assumed that the shivering consumption of the machine wounded person is about 10 , the limb activity consumption is 5 , the basal metabolism is 60 , and the total energy change rate is -75 , which is used to update the energy reserve index.

[0050] Finally, the core body temperature change rate, ventilation rate, hypoxia index and energy change rate calculated above are superimposed with the current simulation state vector to form the simulation state vector of the current time step ; wherein the core body temperature and the energy reserve are superimposed on the corresponding dimension variables of the simulation state vector of the last moment through the core body temperature change rate and the remaining energy change rate of the current moment to form the corresponding dimension variables of the current moment ; is the hypoxia index converted from the hypoxia risk index of the current moment , and is the ventilation rate of the current moment .

[0051] In summary, it can be understood that the physiological mapping module 200 can adaptively update the simulation state according to different water temperatures, oxygen concentrations and water flow conditions by driving the physiological state calculation with real-time environmental data, and jointly participate in mapping by using heat balance, oxygen metabolism and energy consumption to ensure the multidimensional accuracy of physiological state prediction. The finally generated simulation state vector can be continuously iterated for subsequent injury evolution prediction, realizing dynamic injury simulation in time series.

[0052] Regarding the system of the embodiments of the present application, the core objective of the injury evolution prediction module 300 for injury prediction is to predict the evolution trend of the physiological state and injury state of the machine injured person in the future time period according to the current environmental state vector and the simulation state vector, and generate a comprehensive physiological state vector in the future time sequence. It can be understood at first that this module considers both physiological response and simulates injury evolution combined with pathological mechanism, realizing time series prediction of dynamic change of injury in the floating environment.

[0053] Specifically, first, the injury evolution prediction module 300 receives the environmental state vector of the current time step: , and the simulation state vector of the current time step: , In addition to receiving the two types of state vectors of the current time step, considering the relationship between the simulation of injury evolution and the accumulation of physiological factors, such as the physiological change factors of residual energy and hypoxia index, which are the result of historical accumulation, in some embodiments, in addition to the two types of vectors of the current time step, the two types of vectors corresponding to a plurality of historical time steps are also input vectors. After that, the input vectors are standardized to normalize the values of each dimension to a calculable interval to ensure the stability and consistency of the subsequent pathological and dynamic model operations.

[0054] Regarding the calculation of the physiological change rate, including the calculation of the core body temperature change rate, the ventilation rate and the hypoxia change rate and the energy change rate.

[0055] Specifically, first, for the calculation of the core body temperature change rate, it is mainly based on the heat balance model to calculate the change rate of the core body temperature deviating from the standard physiological value which can be directly calculated using the above calculation result. It can be understood that it mainly integrates the average water temperature, temperature gradient, metabolic heat and limb movement heat consumption.

[0056] Regarding the ventilation rate and the hypoxia change rate, it mainly combines the oxygen metabolism model and the oxygen supply and demand balance mechanism to map the ventilation rate and the hypoxia risk into the change rate of the influence of hypoxia on tissue function.

[0057] The rate of change of the remaining energy can also be understood as the time utilization rate, which is specifically the rate of change of the energy reserve based on the remaining energy, the shivering consumption and the activity intensity The rate of change can also be directly calculated using the above calculation results.

[0058] The rate of change of the injury is also divided into core temperature related injury, hypoxia related injury and energy related injury.

[0059] Specifically, the core temperature related injury is to convert the deviation of the core temperature from the standard value into the rate of decline of the tissue function by using the low temperature stress pathological model, and in some embodiments of the present application, the calculation formula is: ; Wherein, is the rate of change of the body temperature related injury, is the standard physiological body temperature, is a pathological mapping function which is constructed according to the specific pathological causes.

[0060] The hypoxia related injury is to calculate the degree of hypoxia by combining the ventilation rate and the oxygen supply and demand balance mechanism, and to map it into the probability of tissue damage. The calculation formula is:

[0061] ; Wherein, is the rate of change of the hypoxia related injury, and conventionally, The function is constructed according to sample experimental data.

[0062] It needs to be understood that the physiological mapping module 200 and the injury evolution prediction module 300, although both involve the use of heat balance model, oxygen metabolism model and energy consumption dynamics model for calculation, their inputs, outputs and mechanisms are different.

[0063] Specifically, the physiological mapping module 200 calculates the instantaneous physiological response of the machine wounded person at this moment under the current environmental conditions, to generate the simulation state vector of the current time step, which is more close to the direct mapping of the current environment and the current physiological response.

[0064] The input of the physiological mapping module 200 is the environmental state vector of the current time step, which uses the heat balance model to calculate the instantaneous body temperature change according to the temperature mean value and gradient, and uses the oxygen metabolism model to calculate the instantaneous ventilation rate and hypoxia risk index according to the dissolved oxygen concentration and hydrodynamic disturbance. The output of the physiological mapping module 200 is the simulation state vector of the current time step, which belongs to the direct physiological response calculation.

[0065] The injury evolution prediction module 300 predicts the next time step of the simulation state, that is, how the injury will evolve in the future, based on the existing historical and current states. It is more inclined to predict the evolution of the state at the next time according to the current physiological state and environmental influence.

[0066] The input of the injury evolution prediction module 300 is the historical and current time step environmental state vector and the historical and current time step simulation state vector, that is, the result sequence calculated by the physiological mapping module 200. The heat balance model, the oxygen metabolism model and the energy consumption dynamics model not only calculate the current change, but also calculate the change rate, such as the body temperature drop rate and the energy consumption rate. The pathological mechanism model converts these change rates into injury evolution trends. For example, the pathological mechanism model judges whether to aggravate the organ function impairment trend according to the calculated hypoxia risk. Finally, the injury evolution prediction module 300 outputs the next time step simulation state vector, which contains the result of the superposition of the physiological state and the injury trend, which belongs to dynamic prediction calculation.

[0067] For example, the physiological mapping module 200 detects that the seawater temperature is 12℃, the dissolved oxygen concentration is 5 , the flow rate is 1.5m / s, and calculates that the body temperature of the machine injured person at the current time is 36.2℃, the respiratory rate is increased to 22 times / min, and the hypoxia risk level is moderate. The injury evolution prediction module 300 predicts that the body temperature will drop to 36.0℃ in the next time step, for example, 10 seconds later, the respiratory rate will further increase to 24 times / min, and the hypoxia risk will upgrade to high, based on the above results, combined with the body temperature drop rate (-0.2℃ / min), the energy consumption rate, etc.

[0068] Finally, regarding the energy-related injury, it is based on the remaining energy, shivering consumption and limb activity intensity to calculate the energy depletion rate and map it to the injury evolution rate. Generally, it is also a corresponding function constructed according to specific pathological causes, which is not specifically limited in different cases.

[0069] It needs to be understood here that the physiological change rate is calculated as the direct output of the physical dynamics and metabolic model, which belongs to environmental and physiological mapping. The injury evolution rate is calculated as the pathological mapping output, which converts the physiological change into the injury index. When iterating, the physiological change rate and the injury evolution rate are superimposed on the current simulation state vector to obtain the next time step state.

[0070] Finally, the physiological change rate is superimposed on the simulation state vector at the current time: ; At the same time, the injury evolution rate is mapped to the injury state component: ; The simulation state vector of the next time step is formed comprehensively: This simulation state vector is then mapped to the corresponding action control of the subsequent machine casualty.

[0071] In addition, in some embodiments of the present application, the simulation state vector of the next time step is taken as the input of the next iteration, combined with the environment state vector of the next time The simulation state vector sequence of the future time steps is iteratively generated: Wherein, is the unit number of time steps.

[0072] From the perspective of system application, the technical point finally realized is the action control mentioned above, that is, the simulation state vector of the next time step is predicted according to the two types of state vectors of the current time step, and then mapped to the corresponding action execution, so that the machine casualty simulates the performance of body temperature drop, breathing difficulty, limb weakness, etc. in water. That is, the control layer only depends on one-step prediction result, which ensures that the action performance is timely and consistent with physiological lag.

[0073] Although the execution action only takes one-step prediction result, the simulation state vector sequence predicted by iteration is mainly used for trend analysis and training optimization model. For example, the prediction sequence of the simulation state vector iteratively generated can see whether the injury is stable, worsens or recovers after 10 minutes, so that it is not clear which to which is only used for real-time control, and it is also helpful for search and rescue evaluation. For example, in search and rescue training, it can be seen in advance that "if the rescue is not continued, the heart may stop beating after 10 minutes", which is more consistent with real risk training.

[0074] Regarding the use of the predicted simulation state vector sequence to optimize the training model, it can be understood that the historical data is compared with the prediction sequence and the actual acquisition result, and the pathological parameters are iteratively optimized, so that the long-term convergent model makes the short-term prediction more accurate.

[0075] Suppose that in the floating environment, the initial core body temperature of the machine casualty is 36.5℃, the ventilation rate is 15 , the hypoxia index is 0.2, the energy reserve , after 30 minutes of iteration, the body temperature drops to 36℃, the ventilation rate is 18 , the hypoxia index rises to 0.45, and the energy reserve drops to 75%, and the corresponding low temperature and hypoxia injury evolution rate is reflected in the comprehensive state vector.

[0076] ​​​​Finally, the execution component, i.e., the operation control module 400, the main function of which is to map the future time sequence of the comprehensive physiological state vector output by the injury evolution prediction module 300 into the physiological performance parameters of the machine wounded person, and further convert them into executable action control instructions, so as to realize the autonomous evolution and action simulation of the machine wounded person in the marine distress environment.

[0077] Specifically, first, the comprehensive physiological state vector at each time step, including core body temperature, ventilation rate, hypoxia index and energy reserve, is mapped into corresponding high-level action parameters by establishing a state-action mapping function (sample experiment establishment). The high-level action parameters include the following four indexes.

[0078] The body temperature adjustment index is generated according to the core body temperature change prediction, and the power adjustment value of the temperature simulator is generated. The floating posture index adjusts the air bag inflation amount and posture control strategy in combination with the energy reserve and physical consumption; The breathing rhythm index mainly generates the output rhythm of the breathing simulator according to the ventilation rate and the hypoxia index. The limb movement amplitude index adjusts the angle and speed of the four-limb driving motor in combination with the energy consumption and injury evolution trend.

[0079] The numerical sequence of each high-level action parameter is output after processing, which is used for subsequent bottom execution mapping. It can be understood that the high-level action parameters are converted into the control amount of the corresponding component to realize physical action and physiological performance.

[0080] Specifically, in some embodiments of the present application, the machine wounded person is provided with: a temperature simulator (head, trunk, limbs), an air bag, a driving motor (limbs) and a breathing simulator.

[0081] The temperature simulator at each place is used to control the heating or cooling power corresponding to the predicted body temperature change.

[0082] The valve opening of the air bag controls the buoyancy and floating posture by adjusting the inflation amount. For example, the valve opening of 10° corresponds to an increase of 5 cm in the floating height compared with the initial height.

[0083] The driving motor control of the limbs controls the angle and speed to realize the limb movement amplitude. For example, the arm motor rotates 30° at a speed of 0.5 rad / s Corresponding to the simulation of waving action, the distress signal is sent.

[0084] The breathing simulator controls the inspiration / expiration cycle according to the breathing rhythm index. For example, 12 breathing cycles per minute correspond to the predicted ventilation rate.

[0085] Herein Figure 3 and Figure 4For the convenience of understanding the execution of each action of the machine casualty, please refer to Table 1 below.

[0086] Table 1: Predicted comprehensive physiological state and corresponding action execution at each time step In summary, it can be understood that the construction of the environmental state vector is to abstract the marine environment in which the machine casualty is located into a vector that can be used for calculation. Specifically, the average water temperature and temperature gradient are extracted according to the data of the water temperature measurement point, the dissolved oxygen concentration in water and the water dynamic disturbance intensity are detected and extracted, the average temperature and gradient reflect the influence of thermal distribution on human body heat loss; the dissolved oxygen and water dynamics reflect the difficulty of breathing and activity. Finally, the four-dimensional vector is integrated. It should be noted that at each time step set, the environmental state vector is recalculated to ensure that the simulation is synchronized with the environmental changes.

[0087] For the construction of the simulation state vector, the physiological state of the machine casualty is represented by a vector to provide a basis for predicting the next state. Specifically, the current time step physiological response is calculated using the environmental state vector obtained above and the heat balance and oxygen metabolism model, the body temperature change rate is obtained using the heat balance formula, the ventilation rate and hypoxia risk are calculated according to the oxygen concentration and water dynamic disturbance, and the energy reserve change is calculated by metabolism, activity consumption and shivering consumption. Finally, the simulation state vector corresponding to the time step is obtained through discretization. In addition, the current time step physiological response is calculated, and the next action is not directly predicted. It can be understood as the real-time state of the machine casualty in the current environment.

[0088] Then, according to the environmental state vector and the simulation state vector of the current time step, and in some embodiments, the states of several time steps in the past can also be considered to improve the prediction stability. According to the environmental state vector and the simulation state vector of the current time step, the change rate of the physiological data in each dimension under the current environment is calculated using the heat balance, oxygen metabolism and energy consumption kinetics model. The current physiological change is mapped to the injury index to predict the physiological and injury changes at the next time step.

[0089] Finally, the predicted next time step simulation state is mapped to the actual action control signal, which is then converted into specific execution component control quantity to drive the machine casualty action. Among them, the body temperature is used to control the heating power of the temperature simulator, the floating posture is controlled by controlling the air bag valve opening degree, the ventilation rhythm is output by the breathing simulator, and the limb movement on the sea is realized by controlling the angle and speed of the built-in drive motor of each torso of the machine casualty. It should be noted that the action executed at the current time step is actually the result of the predicted state calculated at the last time step, and the action at the next time step is recalculated according to the current environment detection.

[0090] The above merely describes preferred embodiments of the present application and is not used to limit the present application, and the patent protection scope of the present application is subject to the claims, and any equivalent structural changes made according to the content of the specification and drawings of the present application should also be included in the protection scope of the present application.

Claims

1. A simulated robotic casualty system based on search and rescue environmental perception, used to control robotic casualties, characterized in that, include: The environment acquisition module is used to acquire environmental information in the water environment area where the machine casualty is located, including dissolved oxygen concentration, hydrodynamic disturbance intensity, and temperature data captured by multiple temperature sensing detection components distributed on the machine casualty. After feature extraction of the environmental information at each time step, the corresponding environmental state vector is constructed. The physiological mapping module receives the environmental state vector in real time, combines it with the human body's thermal balance model and oxygen metabolism model, calculates the machine wounded soldier's instantaneous physiological response in the current environment, and generates a simulated state vector for the current time step; wherein, based on the temperature mean and gradient in the temperature data, the thermal balance model is used to calculate the instantaneous body temperature change; based on the dissolved oxygen concentration and hydrodynamic disturbance index, combined with the oxygen metabolism model, the ventilation rate and hypoxia risk index are calculated. The injury progression prediction module takes the environmental state vectors and simulated state vectors of the historical and current time steps as inputs. Based on the heat balance model, oxygen metabolism model, and energy consumption kinetic model, it calculates the physiological change rates of body temperature, ventilation rate, hypoxia risk, and energy reserves, respectively. Combined with the pathological mechanism model, it predicts the injury progression trend. The physiological change rates and injury progression trend are superimposed on the simulated state vector of the current time step to predict the simulated state vector of the next time step. The operation control module maps the predicted simulated state vector of the next time step to the physiological performance parameters of the machine operator, and converts them into the action execution control signal corresponding to the next time step.

2. The simulated machine casualty system based on search and rescue environmental perception according to claim 1, characterized in that, The environmental acquisition module, through sensing units distributed on and around the machine's in-flight personnel, collects real-time data on the external environmental conditions of the surrounding water environment and analyzes its own surface temperature. Specifically, this includes: The dissolved oxygen concentration in the water body at the current time step is obtained using a dissolved oxygen sensor; Local temperatures are captured by temperature-sensing components placed on the head, torso, and limbs of the injured person on the machine, forming a local temperature sequence. The temperature gradient is calculated and the average temperature is extracted by the temperature difference between the areas of each temperature-sensing component. The acceleration of the water flow around the injured person in the machine is detected by an inertial measurement unit as the intensity of the hydrodynamic disturbance.

3. The simulated machine casualty system based on search and rescue environmental perception according to claim 2, characterized in that, The constructed environmental state vector includes four dimensions: average water temperature, temperature gradient, dissolved oxygen concentration, and hydrodynamic disturbance intensity. These variables are standardized using a feature extraction function to form a unified vector form. ； in, For the current time step Average water temperature For the current time step temperature gradient, For the current time step Dissolved oxygen concentration, For the current time step The intensity of hydrodynamic disturbance.

4. The simulated machine casualty system based on search and rescue environmental perception according to claim 1, characterized in that, The physiological mapping module acquires the environmental state vector output by the environmental acquisition module, and combines it with the human body's thermal balance model and oxygen metabolism model to predict the physiological evolution trend of the injured robot in the current environment, thereby generating the simulated state vector; specifically: Based on the average water temperature and temperature gradient of the aquatic environment, combined with the heat balance model, the rate of change of body temperature in the core area of ​​the machine wounded is calculated. Based on the oxygen concentration and hydrodynamic disturbance intensity of the aquatic environment, combined with the oxygen metabolism model and human respiratory dynamics, the ventilation rate and hypoxia risk index at the current time step are calculated. Based on the tremor expenditure of the human body and the intensity of limb activity, the rate of change of the remaining energy of the injured machine at the current time step is calculated.

5. A simulated machine casualty system based on search and rescue environmental perception according to claim 4, characterized in that, The constructed simulated state vector includes four dimensional variables: body temperature, ventilation rate, hypoxia index, and energy reserve. The simulated state vector takes the following form: ； Among them, the body temperature of the core area was set. and energy reserves Each through the current time step The rates of change of core body temperature and the rates of change of remaining energy are superimposed on the next time step. Based on the corresponding dimension variables of the simulated state vector, the prediction forms the next time step. The corresponding dimension variable; For the current time step The hypoxia risk index is converted into a hypoxia indicator. For the current time step ventilation rate.

6. A simulated machine casualty system based on search and rescue environmental perception according to claim 5, characterized in that, The injury evolution prediction module receives the environmental state vector and simulated state vector at the current time step and a set number of historical time steps, standardizes the input vectors, and calculates the injury evolution trend using a pathological model; specifically including: Based on the thermal balance model, the deviation between the core body temperature and the standard physiological value is calculated, and the deviation is converted into the rate of tissue function decline using the low temperature stress pathological model. By combining ventilation rate, hypoxia risk index and oxygen supply and demand balance mechanism, the degree of hypoxia is calculated and mapped to the probability of injury using pathological mechanism. By combining remaining energy, activity intensity, and shivering expenditure, the energy depletion rate is calculated and mapped to the energy-related damage evolution rate through a pathological model.

7. A simulated robotic casualty system based on search and rescue environmental perception according to claim 6, characterized in that, The physiological rate of change is superimposed onto the simulated state vector at the current time step: ； Simultaneously, the trend of injury evolution is mapped into injury status components: ； Combined, the simulated state vector for the next time step is formed: 。 8. A simulated robotic casualty system based on search and rescue environmental perception as described in claim 1, characterized in that, The operation control module maps the simulated state vector at each time step to high-altitude motion parameters, including adjusting body temperature, floating posture, breathing rhythm, and adjusting limb movement amplitude.

9. A simulated robotic casualty system based on search and rescue environmental perception according to claim 8, characterized in that, The high-level action parameters are converted into low-level execution instructions, which correspond to the control quantities of specific execution components; specifically including: Control the heating and cooling power of the temperature-sensing simulator installed inside the machine to correspond to changes in body temperature; Adjust the valve opening of the airbag installed inside the machine wounded to control buoyancy and floating height; The angle and speed of the drive motors installed in the limbs of the injured person are controlled separately, corresponding to the range of limb movement; Control the output of the breathing simulator set inside the machine wounded, corresponding to the ventilation rhythm.

Images