Multifunctional integrated vehicle safety control method and system based on UWB

Abstract

The application discloses a multifunctional integrated vehicle-mounted safety control method and system based on UWB, which comprises the following steps: collecting dynamic environment data around and in the vehicle based on multiple sensors in real time, adjusting the path and optimizing the driving mode by generating a multi-level three-dimensional environment model, using the external data such as obstacles, road conditions and weather collected by the sensors, combining the behavior mode of the driver and the health state of the passengers to perform risk assessment and dynamic adjustment, and gradually activating the protection system by analyzing the vehicle speed, acceleration and other factors in the collision prediction and protection mechanism to ensure safety, optimizing the energy distribution of key systems by the energy scheduling model to improve energy efficiency, and preferentially ensuring the operation of the safety system in an emergency. The application integrates an adaptive learning algorithm, optimizes the driving assistance function in multiple scenes, and realizes adaptive adjustment and safety control of the vehicle in different driving environments.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle safety control, in particular to a multi-functional integrated vehicle safety control method and system based on UWB. BACKGROUND

[0002] With the rapid development of intelligent driving and autonomous driving technology, vehicle safety control systems have gradually become an important technical means to improve vehicle safety and driving performance. In the prior art, vehicles mainly rely on cameras, radars, ultrasonic sensors and other sensors to collect environmental information around the vehicle, and adjust the driving path, speed and braking in combination with the driver's operation. These technologies have the problems of limited perception range, insufficient accuracy and poor adaptability to complex environments. Especially in a dynamically changing driving environment, traditional sensor systems are difficult to capture rapidly changing external information in real time, resulting in reduced efficiency and safety of driving assistance functions.

[0003] Although the existing vehicle safety control system has improved the safety and driving experience of the vehicle to some extent, there are still many technical defects. The current vehicle safety control system mainly relies on cameras, radars, ultrasonic sensors and other traditional environmental perception devices. These devices can provide external environmental information within a certain range, but their perception ability is insufficient in complex and variable driving environments. Although cameras can recognize obstacles and vehicles, their performance will decrease significantly at night or in bad weather conditions, resulting in a blind area. Although radars and ultrasonic sensors can capture obstacles at close range, they perform poorly at long distances, especially in high-speed driving scenarios. The response time of radar sensors is often not fast enough. Due to these limitations, the existing system cannot accurately perceive and predict potential dangers in complex driving environments, resulting in the vehicle being unable to take effective measures in time, thereby increasing the risk of driving.

[0004] The existing vehicle safety system also has certain limitations in risk assessment. Most systems can only perform risk analysis based on single-dimensional data, lacking comprehensive evaluation capability of multi-dimensional data. This single risk judgment method ignores the influence of multiple important factors, such as the complexity of the road, the dynamic changes of obstacles, and the behavior patterns of external targets. The risk assessment of the existing system is usually static and difficult to dynamically adjust according to changes in external environment and vehicle state. This risk assessment method leads to the system being unable to make accurate risk predictions when facing variable driving environments, increasing the likelihood of accidents.

[0005] In terms of collision prediction, the prior art usually relies on simple preset algorithms or rules, which makes the accuracy of collision prediction low. Since the existing system only relies on the speed of the vehicle and the distance of the obstacle to predict the potential collision risk, it cannot consider the influence of multi-dimensional factors, leading to misjudgment or omission of the system in complex driving environment. The existing system has a long reaction time when a potential collision occurs, especially in high-speed or emergency situations, the system is difficult to make a quick judgment and activate the protection measures, increasing the probability of collision and reducing the safety of the vehicle.

[0006] The existing energy management system of the vehicle also has many deficiencies. Current energy management usually relies on fixed energy distribution strategies, which cannot be flexibly adjusted according to the real-time needs of the vehicle and environmental changes. The existing system cannot dynamically adjust the distribution of vehicle energy when the driver's operation and external environment change, resulting in low energy utilization efficiency. In emergency situations, the existing energy management system also lacks the ability to respond, and cannot prioritize energy support for key safety systems such as braking and steering, which seriously affects the safety and emergency response capability of the vehicle. The fixed energy distribution strategy not only makes it difficult to adapt to complex driving scenarios, but also easily causes energy waste, which cannot meet the energy saving requirements.

[0007] In terms of passenger health monitoring, although some high-end vehicles have introduced health monitoring functions, these functions are mostly independent of other systems of the vehicle and cannot effectively link with the driving control system. Although the system can monitor physiological data such as heart rate and respiratory rate of the passengers, these monitoring data can only be used for warning and cannot directly affect the driving mode of the vehicle. The existing health monitoring system lacks linkage with the driving state and external environment, and cannot adjust the driving mode in a timely manner according to the health status of the passengers. Especially when the passenger's health is abnormal, the system cannot quickly respond and adjust the vehicle speed or switch to automatic driving mode, resulting in potential safety risks.

[0008] Therefore, how to provide a multi-functional integrated vehicle safety control method and system based on UWB is a problem that needs to be solved by those skilled in the art. SUMMARY

[0009] One object of the present application is to provide a multi-functional integrated vehicle safety control method and system based on UWB. The present application configures a UWB sensor network, combines an adaptive learning algorithm, generates a multi-level three-dimensional environment model, and monitors and analyzes the dynamic environment, external targets and passenger health status in real time during vehicle driving. The method describes in detail the operation steps of risk prediction, path adjustment, collision protection and energy scheduling based on a multi-factor weight model, and has the advantages of high real-time sensing accuracy, accurate risk prediction, energy efficiency optimization and strong passenger safety guarantee.

[0010] The UWB-based multifunctional integrated vehicle safety control method and system according to the embodiment of the present application comprises the following steps:

[0011] S1, collecting multi-level dynamic environment data around the vehicle and in-vehicle state information in real time through the UWB sensor network arranged around the vehicle, and automatically adjusting the sensing range, sensitivity and frequency of the sensor according to the dynamic complexity of the external environment and the current driving state of the vehicle, and generating a multi-level three-dimensional environment model through a self-adaptive learning algorithm;

[0012] S2, based on the generated multi-level three-dimensional environment model, predicting and evaluating potential risk events by introducing a multi-factor weight model of time priority, space priority, speed priority and dynamic behavior prediction, and automatically generating a priority list containing real-time risk events according to the changes of the environment and the vehicle state and dynamically adjusting the risk weight parameters;

[0013] S3, based on the generated priority list, the system obtains the dynamic path and behavior intention data of the external target through UWB communication with external devices, combines the path adjustment of the vehicle with the behavior prediction result of the external target, and intelligently adjusts the driving path, speed and driving mode through fusion with the behavior mode database of the driver;

[0014] S4, the system actively adjusts the driving mode of the vehicle based on the real-time monitoring of the passenger health state data and the combination of path adjustment and external environment changes, and when the passenger health data is abnormal, the system preferentially switches to low-speed and automatic driving mode according to the environmental complexity and the results of dynamic adjustment of the vehicle path;

[0015] S5, combining the generated three-dimensional environment model and path adjustment data, predicting the energy demand according to the current environment, driving state and future driving path, and optimizing the energy distribution of the vehicle in real time through an adaptive energy scheduling model, and the system pre-plans the energy consumption priority in different scenarios according to external factors, and preferentially guarantees the energy supply of key equipment in emergency situations;

[0016] S6, the system predicts the complex collision scenarios based on the generated three-dimensional environment model and path adjustment information using a multi-dimensional collision prediction model, and the system predicts the potential collision type by analyzing the relative position, speed, acceleration and deceleration trend and direction change of the vehicle, obstacles and external targets, and gradually activates the adaptive multi-level protection mechanism in stages according to the severity and probability of the collision prediction;

[0017] S7, based on the driver's behavior pattern library, the system analyzes the current state of the driver and the external environment through the behavior pattern fusion mechanism, when the driver approaches the vehicle, the system automatically matches the behavior pattern and adjusts the settings in the vehicle, and optimizes the vehicle starting process and provides a seamless entry experience according to the current energy state of the vehicle and the behavior pattern of the driver;

[0018] S8, through multi-scene adaptive learning algorithm, multi-objective optimization adjustment is carried out on the driving assistance function in each scene, combining with the change of external environment, driver operation mode and vehicle performance, the acceleration, steering and braking response is optimized, in similar environment, the system further optimizes the driving strategy based on the previous optimization result, realizes the adaptive multi-scene learning of vehicle.

[0019] Optionally, the S1 specifically comprises:

[0020] S11, through the multiple UWB sensors arranged around the vehicle, the multi-level dynamic environment data around the vehicle and the in-vehicle state information are collected in real time, the environment data includes the spatial position, relative speed, terrain change, weather condition and traffic flow density of the obstacle, and the data is classified into static environment data and dynamic environment data;

[0021] S12, based on the collected static and dynamic environment data, the real-time evaluation of environment complexity is carried out, and the sampling frequency and accuracy of the sensor are adjusted in real time to adapt to the complex change of the environment:

[0022]

[0023] Wherein, C e represents the environment complexity, T represents the traffic flow density, W represents the influence of weather condition, V is the current speed of the vehicle, represents the relative movement change of the obstacle, and α and β are adjustment factors, and ∈ is a stable term;

[0024] S13, based on the evaluation result of the environment complexity, the system dynamically adjusts the data weight of each sensor through the adaptive learning algorithm:

[0025]

[0026] Wherein, is the weight of the sensor data, δ is the adaptive learning rate, is the weight adjustment result according to the environment complexity, γ represents the weight factor of the obstacle change, and η is the balance coefficient;

[0027] S14, after adjusting the sensor weight, the system generates a multi-level three-dimensional environment model according to the weighted sensor data, the model includes the position of dynamic obstacles, the slope change of terrain, the traffic flow density and its change trend, according to the real-time data and the change of sensor weight, the three-dimensional environment model is continuously updated:

[0028]

[0029] Wherein, M 3D is a three-dimensional environment model, P i (t) represents the data packet collected by the sensor at time t, W s (t) is the sensor weight at the time.

[0030] Optionally, the S2 specifically comprises:

[0031] S21, the system obtains factor information related to time, space and speed based on the generated multi-level three-dimensional environment model;

[0032] S22, a multi-factor weight model of time priority, space priority and speed priority is introduced, which is used for weighted calculation of time, space and speed factors:

[0033]

[0034] Wherein, R w is a risk weight, T f is a time factor, S f is a space factor, V f is a speed factor, and α, β and γ are weight coefficients, and Δt is a time change;

[0035] S23, on the basis of the weight model, the system introduces a dynamic behavior prediction factor for predicting the behavior path of the external target, and uses the motion information of the external target, combines historical data and real-time data to predict its future behavior:

[0036]

[0037] Wherein, P d represents a dynamic behavior prediction value, O i (t) is the moving track of the external target at time t, V t is the current speed of the vehicle, and S i is the distance between the target and the vehicle;

[0038] S24, the system generates a priority list of risk events according to the time, space, speed factors and dynamic behavior prediction value P d , through priority list sorting, high priority events are processed first, and low priority events enter the waiting queue.

[0039] Optionally, the S3 specifically includes:

[0040] S31, the system communicates with the external device through UWB according to the generated priority list to obtain the dynamic path and behavior intention data of the external target, and the external target data includes the relative position, speed, direction and behavior prediction of the target;

[0041] S32, the path adjustment and external behavior prediction combined system adjusts the path according to the dynamic behavior prediction value of the external target and the driving state of the vehicle:

[0042]

[0043] Wherein, P adj is the path adjustment amount, V t is the current speed of the vehicle, S o is the relative distance of the external target, P d is the dynamic behavior prediction value, λ1 is the adjustment coefficient, T r is the response time, μ1 is the prediction weight, and ∈ is the stability term;

[0044] S33, the system combines the behavior prediction of the external target with the historical behavior pattern of the driver to optimize the path, speed and driving mode of the vehicle, and intelligently adjusts according to the behavior pattern of the driver and the change of the current external environment;

[0045] S34, based on the path adjustment P adj and intelligent adjustment, dynamically adjust the driving speed and driving mode of the vehicle, and the adjustment of speed and path is linked to make the vehicle maintain the safety and stability of driving in complex external environment, and through real-time analysis of external target data and the behavior pattern of the driver, the optimal driving mode and speed adjustment scheme is selected.

[0046] Optionally, the S4 specifically includes:

[0047] S41, the health status of the passenger is monitored in real time through the sensor in the vehicle, including heart rate, respiratory rate, body temperature and posture change, and any abnormal situation is identified according to the real-time analysis of the health status data;

[0048] S42, the obtained health status data is combined with the current driving path and external environment condition, and if the health status of the passenger is abnormal, the system determines whether to adjust the driving mode by comprehensively considering the current driving path and weather condition;

[0049] S43, the system judges whether the speed and path of the vehicle need to be adjusted according to the passenger health data and the path adjustment amount P adj ​

[0050]

[0051] wherein, H adj is the adjustment amount based on the health status, H s represents the change in the passenger's health status, P adj is the path adjustment amount, and θ is the adjustment coefficient, T h is the time interval for health data collection, and ∈ h is a stability term;

[0052] S44, when the system detects an abnormal passenger health status, it automatically switches to a low-speed mode or an autonomous driving mode according to road complexity and weather conditions, and the switching of the driving mode is based on the urgency of the passenger's health status and the risk assessment result of the external environment;

[0053] S45, by continuously updating the three-dimensional environment model through real-time monitoring of the passenger's health status and combining the UWB communication of external devices, the system further optimizes the driving mode and path adjustment based on the updated environment model, so that the passenger's health status matches the driving environment, and provides a real-time response driving scheme.

[0054] Optionally, the S5 specifically includes:

[0055] S51, based on the current driving path, external environment and vehicle operating state, the energy demand in different driving scenarios is predicted, and the real-time road conditions, driving speed and vehicle load are combined to match the energy demand with the actual driving scenario, and the system adjusts the energy prediction value in real time according to the changing environmental conditions to optimize energy utilization;

[0056] S52, the energy demand is dynamically adjusted by real-time monitoring of external environmental data, and after the external factors are combined with the energy demand, the energy distribution scheme can be adjusted according to the complexity of different driving scenarios to maximize vehicle energy efficiency and driving performance;

[0057] S53, the system dynamically optimizes the energy distribution of the vehicle using an adaptive energy scheduling model based on the energy demand prediction result and external environmental conditions, and adjusts the energy distribution of each key system in real time, and optimizes the energy distribution strategy through the built-in self-learning algorithm according to the driving habits and historical driving data of the driver;

[0058] S54, when the passenger's health status is detected to be abnormal, the weather is bad, or there is an emergency traffic accident, the system preferentially allocates energy to the key safety system, and in an emergency, the system will adjust the energy priority according to the risk level of the sudden event to ensure that the key system can still work normally in the case of energy shortage;

[0059] S55, the system can provide real-time feedback according to external data in extreme environments, further optimize energy demand, and in extremely low temperature environments, the system prioritizes energy for critical systems and adjusts energy consumption according to real-time feedback to adapt to low temperature and high energy consumption special scenarios;

[0060] S56, the system continuously monitors the energy consumption data of the vehicle and external feedback information, and optimizes the energy scheduling strategy through real-time adaptive learning algorithm, the feedback mechanism allows the system to quickly respond when the external environment changes, dynamically adjusts the energy distribution, so that the vehicle can realize energy efficiency optimization and ensure driving safety in complex and sudden environments.

[0061] Optionally, the S6 specifically includes:

[0062] S61, by collecting the multi-dimensional information of the speed, acceleration, relative position and direction change of the external target of the vehicle, the system constructs a collision prediction model to calculate the potential collision scenario, and the system updates the collision prediction in real time according to the multi-dimensional information:

[0063] C pred = f(V t ,a t ,O d ,Δθ);

[0064] Wherein, C pred is the collision prediction value, V t represents the current speed of the vehicle, a t is the current acceleration, O d is the relative distance of the external obstacle, and Δθ is the direction change of the vehicle and the obstacle;

[0065] S62, based on the collision prediction result, the system calculates the potential collision risk level, and the system adjusts the risk level according to the current driving state of the vehicle and external environmental factors:

[0066]

[0067] Wherein, R level is the collision risk level, W e is the environmental influence factor, E s is the vehicle state influence factor, and ∈ r is the stability coefficient;

[0068] S63, when the system detects a higher risk of collision, the system gradually activates the protection mechanism according to the risk level, including early warning and deceleration prompt, activates higher level protection system, realizes the activation of each protection measure in turn, and protects the safety of the vehicle and passengers;

[0069] S64, after activating the protection mechanism, the system continues to monitor the speed, acceleration and dynamic information of external obstacles of the vehicle, and adjusts the protection measures according to the real-time collision prediction results, so as to realize the optimization of the protection scheme with the change of external environment;

[0070] S65, the system prioritizes the allocation of energy to the key protection system through linkage with the energy management system, and if the vehicle is in the scene of energy shortage and high energy consumption, the system will automatically adjust the energy allocation scheme to make the key protection work normally;

[0071] S66, when the system activates the protection mechanism, it dynamically adjusts in combination with the operation of the driver, and the system optimizes the protection scheme according to the operation input of the driver, so that the operation of the driver is properly responded, and the execution effect of the protection mechanism is optimized.

[0072] Optionally, the S8 specifically comprises:

[0073] S81, the acceleration, braking and steering response data of the vehicle and the external environment data are collected according to different driving environments, and a response model under different driving scenes is constructed to reflect the different influences of each scene on the behavior of the vehicle;

[0074] S82, through an adaptive learning algorithm, the acceleration, braking and steering strategies of the vehicle are dynamically adjusted based on the collected multi-scene data, and the system optimizes the response behavior of the vehicle according to the historical operation mode of the driver and the real-time external environment change:

[0075] R opt =f(V t ,A s ,T r ,E e );

[0076] Wherein, R opt is the optimized vehicle response, V t is the current speed of the vehicle, A s is the acceleration parameter, T r is the steering response time, and E e is the external environment factor;

[0077] S83, the system adjusts the driving strategy in different scenes according to the optimization result, in the urban road scene, the low-speed steering control is given priority, on the highway, the system adjusts the balance of acceleration and braking, and in the mountain road, the linkage control of steering and braking is emphasized, and the system dynamically adjusts the response speed and intensity of acceleration, braking and steering according to the specific scene;

[0078] S84, the system updates the model according to the learning result by combining historical data and real-time data, and continuously updates the adaptive learning model:

[0079] M upd = M prev + α × ΔR;

[0080] wherein, M upd is the updated model, M prev is the previous model parameter, α is the learning rate, and ΔR is the response adjustment amount;

[0081] S85, by real-time monitoring of the driver's operation and the vehicle's running state, adjusting the acceleration, braking and steering strategy of the vehicle according to the driver's operation frequency, road complexity and vehicle dynamics, and adjusting the response of the vehicle at any time;

[0082] S86, the system improves the global driving strategy in multiple scenarios through long-term driving data accumulation. Whenever the vehicle runs in a similar driving scenario, the system optimizes based on historical data and adjusts the strategy in the future in the same and similar scenarios, and optimizes the global driving strategy through long-term data accumulation.

[0083] Optionally, the following modules are included:

[0084] UWB sensor network module: real-time collection of multi-level dynamic environment data around the vehicle and in-vehicle state information, adjustment of the perception range, sensitivity and collection frequency according to the dynamic complexity of the external environment and the driving state of the vehicle;

[0085] Adaptive learning algorithm module: based on the collected data, generate a multi-level three-dimensional environment model, and dynamically adjust the sensor weights to optimize the generation and update of the model, and provide real-time environmental feedback;

[0086] Multi-factor weight evaluation module: through the multi-factor weight model of time priority, space priority and speed priority, predict potential risks, dynamically adjust the weight parameters, and generate a priority list containing real-time risk events;

[0087] Behavior prediction and path adjustment module: according to the dynamic path and behavior intention of external targets, combined with the driver's behavior pattern and the vehicle state, adjust the vehicle driving path, speed and driving mode to realize intelligent adjustment;

[0088] Passenger health monitoring module: monitor the health status of passengers through in-vehicle sensors, combine the driving path and environmental changes, and automatically adjust the driving mode. When abnormal health data is detected, switch to low-speed or automatic driving mode;

[0089] Energy scheduling module: according to the energy demand prediction results and external environmental conditions, real-time optimization of energy distribution of key systems, and in emergency situations, priority is given to energy allocation to safety systems;

[0090] Collision prediction and prevention module: based on a multi-dimensional collision prediction model, real-time evaluation of the relative position, speed, acceleration of the vehicle and obstacles and external targets, and activation of a step-by-step prevention mechanism to deal with different collision risk levels;

[0091] Driving assistance optimization module: based on a multi-scenario adaptive learning algorithm, optimizing acceleration, braking and steering response strategies, adjusting global driving strategies through long-term data accumulation, and optimizing vehicle driving performance in different driving environments

[0092] The beneficial effects of the present application are:

[0093] The present application adopts a multi-level dynamic environment data acquisition technology of UWB sensor network, realizes high-precision perception of the environment around the vehicle, compared with traditional sensors, not only can accurately capture obstacles, terrain changes and traffic flow in complex and dynamic driving environments, but also can adjust the perception range and accuracy in real time according to the current driving state of the vehicle, this dynamic perception method makes the vehicle can quickly adapt to the changes of external environment, significantly improves the accuracy and response speed of perception; The present application predicts and evaluates risk events by combining time, space, speed and other multi-dimensional factors through a multi-factor weight model, this model can dynamically adjust the weight of each factor, generate a priority list of risk events, and provide accurate risk prediction and evaluation results for the system, especially by fusing external target behavior prediction and driver historical behavior pattern, the system can intelligently adjust the driving path and driving mode, realize more accurate path planning and safety optimization, overcome the defects of single risk assessment and insufficient dynamic response capability in the prior art.

[0094] The present application can accurately predict potential collision scenarios by considering vehicle speed, acceleration, relative position and obstacle dynamic behavior through a multi-dimensional collision prediction model, according to the prediction result, the system gradually activates a multi-level prevention mechanism, and starts the prevention measures in stages, from preliminary warning and speed reduction prompt, to higher level prevention system activation, to ensure the safety of the vehicle and passengers in complex road conditions, an adaptive energy scheduling model is introduced, which can dynamically optimize the allocation of vehicle energy according to different driving scenarios and energy demand, by real-time monitoring of external environmental conditions, the system can adjust the energy allocation scheme, especially in emergency, priority is given to the energy support of key safety systems to ensure the safety performance of the vehicle, this adaptive energy management greatly improves the energy efficiency of the vehicle, solves the problem of fixed allocation and insufficient flexibility of traditional energy management system,

[0095] The application can automatically adjust the driving mode by combining the driving path and external environmental changes by monitoring the health state data of the passengers in real time, and when the health state of the passengers is abnormal, the system automatically switches to a low-speed or automatic driving mode according to the environmental complexity and risk assessment results, ensuring the safety of the passengers, which makes up for the insufficient linkage between the health monitoring system and the driving mode in the prior art, improves the safety of the passengers, and through the multi-scene adaptive learning algorithm, the application can continuously optimize the acceleration, braking and steering response strategies of the vehicle in different driving environments, and the system can continuously optimize the global driving strategy through long-term data accumulation and real-time environmental feedback, improving the adaptability of the vehicle in complex environments, which significantly improves the driving performance of the vehicle and makes it perform the best response effect in different driving scenes. BRIEF DESCRIPTION OF DRAWINGS

[0096] The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application, and do not limit the application. In the drawings:

[0097] Fig. 1 The flowchart of the multifunctional integrated vehicle safety control method and system based on UWB proposed by the application;

[0098] Fig. 2 The flowchart of the risk event prediction and assessment and path adjustment based on the multi-factor weight model in the application;

[0099] Fig. 3 The schematic diagram of the key system energy distribution of the adaptive energy scheduling model in the application. DETAILED DESCRIPTION

[0100] The application will now be described in further detail with reference to the drawings. These drawings are simplified schematic diagrams that only schematically illustrate the basic structure of the application, and therefore only show the components related to the application.

[0101] REFERENCE Figs. 1-3 The multifunctional integrated vehicle safety control method and system based on UWB, comprising the following steps:

[0102] S1, through the UWB sensor network configured around the vehicle, real-time collection of multi-level dynamic environmental data around the vehicle and in-vehicle state information, the sensor automatically adjusts the sensing range, sensitivity and frequency according to the dynamic complexity of the external environment and the current driving state of the vehicle, and generates a multi-level three-dimensional environmental model through an adaptive learning algorithm;

[0103] S2, based on the generated multi-level three-dimensional environment model, the potential risk events are predicted and evaluated by introducing a multi-factor weight model of time priority, space priority, speed priority and dynamic behavior prediction, the model dynamically adjusts the risk weight parameters according to the changes of environment and vehicle state, and automatically generates a priority list containing real-time risk events;

[0104] S3, based on the generated priority list, the system obtains the dynamic path and behavior intention data of the external target through UWB communication with external devices, the system combines the path adjustment of the vehicle with the behavior prediction result of the external target, and fuses with the behavior mode database of the driver to intelligently adjust the driving path, speed and driving mode;

[0105] S4, the system actively adjusts the driving mode of the vehicle based on the real-time monitoring of passenger health state data and the combination of path adjustment and external environment changes, when the passenger health data is abnormal, the system switches to low speed and automatic driving mode according to the environmental complexity and the results of vehicle path dynamic adjustment;

[0106] S5, combining the generated three-dimensional environment model and path adjustment data, the energy demand is predicted according to the current environment, driving state and future driving path, and the energy distribution of the vehicle is optimized in real time through an adaptive energy scheduling model, the system pre-plans the energy consumption priority in different scenarios according to external factors, and prioritizes the energy supply of key equipment in emergency situations;

[0107] S6, the system uses a multi-dimensional collision prediction model to predict complex collision scenarios based on the generated three-dimensional environment model and path adjustment information, the system predicts the potential collision type by analyzing the relative position, speed, acceleration and deceleration trend and direction change of the vehicle, obstacles and external targets, and according to the severity and probability of collision prediction, gradually activates the adaptive multi-stage protection mechanism in stages;

[0108] S7, the system analyzes the current state of the driver and the external environment condition through the behavior mode fusion mechanism based on the driver's behavior mode library, when the driver approaches the vehicle, the system automatically matches the behavior mode and adjusts the settings in the vehicle, and according to the current energy state of the vehicle and the behavior mode of the driver, optimizes the vehicle start process and provides a seamless entry experience;

[0109] S8, through a multi-scene adaptive learning algorithm, the driving assistance functions in each scene are optimized and adjusted, combined with external environment changes, driver operation mode and vehicle performance, the acceleration, steering and braking response is optimized, in similar environment, the system further optimizes the driving strategy based on the previous optimization results, realizing adaptive multi-scene learning of the vehicle.

[0110] In this embodiment, S1 specifically includes:

[0111] S11, through a plurality of UWB sensors arranged around the vehicle, real-time collection of multi-level dynamic environment data around the vehicle and in-vehicle state information, the environment data including the spatial position, relative speed, terrain change, weather condition and traffic flow density of obstacles, and the data is classified into static environment data and dynamic environment data;

[0112] S12, based on the collected static and dynamic environment data, real-time evaluation of the environment complexity, real-time adjustment of the sampling frequency and accuracy of the sensors to adapt to the complex changes of the environment:

[0113]

[0114] wherein C e represents the environment complexity, T represents the traffic flow density, W represents the influence of weather conditions, V is the current speed of the vehicle, represents the relative movement change of the obstacle, and a and β are adjustment factors, and ∈ is a stability term;

[0115] S13, based on the evaluation result of the environment complexity, the system dynamically adjusts the data weight of each sensor through an adaptive learning algorithm:

[0116]

[0117] wherein, is the weight of the sensor data, δ is the adaptive learning rate, is the weight adjustment result according to the environment complexity, γ represents the weight factor of the obstacle change, and η is the balance coefficient;

[0118] S14, after adjusting the sensor weight, the system generates a multi-level three-dimensional environment model according to the weighted sensor data, the model including the position of the dynamic obstacle, the slope change of the terrain, the traffic flow density and its change trend, and the three-dimensional environment model is continuously updated according to the real-time data and the change of the sensor weight:

[0119]

[0120] wherein M 3D is a three-dimensional environment model, P i (t) represents the data packet collected by the sensor at time t, W s (t) is the sensor weight at the time.

[0121] In this embodiment, S2 specifically includes:

[0122] S21, the system obtains factor information related to time, space and speed based on the generated multi-level three-dimensional environment model;

[0123] S22, a multi-factor weight model prioritizing time, space and speed is introduced for weighted calculation of the time, space and speed factors:

[0124]

[0125] wherein R w is the risk weight, T f is the time factor, S f is the space factor, V f is the speed factor, and a, b and g are weight coefficients, and At is the time change;

[0126] S23, based on the weight model, a dynamic behavior prediction factor is introduced for predicting the behavior path of the external target, and the future behavior of the external target is predicted by using the motion information of the external target, combining historical data and real-time data:

[0127]

[0128] wherein P d represents the dynamic behavior prediction value, O i (t) is the moving track of the external target at time t, V t is the current speed of the vehicle, and S i is the distance between the target and the vehicle;

[0129] S24, the system generates a priority list of risk events according to the time, space, speed factors and the dynamic behavior prediction value P d , and through the priority list sorting, the high-priority events are processed first, and the low-priority events enter the waiting queue.

[0130] In the embodiment, the S3 specifically comprises:

[0131] S31, the system obtains the dynamic path and behavior intention data of the external target by communicating with the external device through UWB according to the generated priority list, and the external target data includes the relative position, speed, direction and behavior prediction of the target;

[0132] S32, the path adjustment and external behavior prediction combined system adjusts the path according to the dynamic behavior prediction value of the external target and the driving state of the vehicle:

[0133]

[0134] wherein P adj is the path adjustment amount, V t is the current speed of the vehicle, and So P is the relative distance of the external target d λ1 is the adjustment coefficient, T is the dynamic behavior prediction value r μ1 is the prediction weight, ∈ is the stability term

[0135] S33, the system combines the behavior prediction of the external target with the historical behavior pattern of the driver, optimizes the path, speed and driving mode of the vehicle, and intelligently adjusts the behavior pattern of the driver and the changes in the current external environment

[0136] S34, based on the path adjustment P adj and intelligent adjustment, dynamically adjusting the driving speed and driving mode of the vehicle, the adjustment of speed and path is linked, so that the vehicle can maintain the safety and stability of driving in complex external environment, by analyzing the external target data and the behavior pattern of the driver in real time, selecting the optimal driving mode and speed adjustment scheme.

[0137] In this embodiment, S4 specifically includes:

[0138] S41, the health status of the passenger is monitored in real time by the sensor in the vehicle, including heart rate, breathing rate, body temperature and posture change, and real-time analysis is performed according to the health status data, and any abnormal situation is identified

[0139] S42, the obtained health status data is combined with the current driving path and external environment condition, if the health status of the passenger is abnormal, the system determines whether to adjust the driving mode by comprehensively considering the current driving path and weather condition

[0140] S43, the system determines whether the speed and path of the vehicle need to be adjusted according to the passenger health data and the path adjustment amount P adj

[0141]

[0142] H adj is the adjustment amount based on the health status, H s represents the change of the health status of the passenger, P adj is the path adjustment amount, θ is the adjustment coefficient, T h is the time interval of health data collection, ∈ h is the stability term

[0143] S44, when the system detects that the health status of the passenger is abnormal, it automatically switches to the low-speed mode or the automatic driving mode according to the road complexity and the weather condition, the switching of the driving mode is based on the emergency degree of the health status of the passenger and the risk assessment result of the external environment

[0144] ​S45, continuously update the three-dimensional environment model by real-time monitoring of passenger health status and combining UWB communication of external devices, based on the updated environment model, the system further optimizes the driving mode and path adjustment to match the passenger health status with the driving environment, and provides a real-time response driving scheme.

[0145] In this embodiment, S5 specifically includes:

[0146] S51, based on the current driving path, external environment and running state of the vehicle, predict the energy demand in different driving scenarios, combine real-time road conditions, driving speed and vehicle load, match the energy demand with the actual driving scenario, and the system adjusts the energy prediction value in real time according to the changing environmental conditions to optimize energy utilization;

[0147] S52, dynamically adjust the energy demand by real-time monitoring of external environment data, after combining external factors and energy demand, according to the complexity of different driving scenarios, adjust the energy distribution scheme to maximize vehicle energy efficiency and driving performance;

[0148] S53, the system dynamically optimizes the energy distribution of the vehicle according to the energy demand prediction result and external environmental conditions, using an adaptive energy scheduling model, adjusts the energy distribution of each key system in real time, and optimizes the energy distribution strategy according to the driving habits and historical driving data of the driver through the built-in self-learning algorithm;

[0149] S54, when detecting abnormal passenger health status, severe weather or traffic accident emergency, the system preferentially allocates energy to the key safety system, in emergency, the system will adjust the energy priority according to the risk level of the emergency event, so that the key system can still maintain normal work under the condition of energy shortage;

[0150] S55, the system can further optimize the energy demand according to the real-time feedback of external data in extreme environment, in extreme low temperature environment, the system preferentially provides energy for the key system, and adjusts the energy consumption according to the real-time feedback to adapt to the special scene of low temperature and high energy consumption;

[0151] S56, the system continuously monitors the energy consumption data of the vehicle and external feedback information, optimizes the energy scheduling strategy through real-time self-adaptive learning algorithm, and the feedback mechanism allows the system to quickly respond when the external environment changes, dynamically adjusts the energy distribution, so that the vehicle can realize energy efficiency optimization and ensure driving safety in complex and emergency environment.

[0152] In this embodiment, S6 specifically includes:

[0153] S61, by collecting the speed, acceleration, relative position and direction change of the external target of the vehicle, the system constructs a collision prediction model to calculate the potential collision scenario, and the system updates the collision prediction in real time according to the multi-dimensional information:

[0154] C pred = f(V t ,a t ,O d ,Δθ);

[0155] Where C pred is the collision prediction value, V t represents the current speed of the vehicle, a t is the current acceleration, O d is the relative distance of the external obstacle, and Δθ is the direction change of the vehicle and the obstacle.

[0156] S62, based on the collision prediction result, the system calculates the potential collision risk level, and the system adjusts the risk level according to the current driving state of the vehicle and the external environmental factors:

[0157]

[0158] Where R level is the collision risk level, W e is the environmental influence factor, E s is the vehicle state influence factor, and ∈ r is the stability coefficient.

[0159] S63, when the system detects a higher risk of collision, the system gradually activates the protection mechanism according to the risk level, including the initial warning and deceleration prompt, activates the higher level protection system, and realizes the sequential activation of each protection measure to ensure the safety of the vehicle and passengers;

[0160] S64, after activating the protection mechanism, the system continuously monitors the speed, acceleration and dynamic information of the external obstacle of the vehicle, and adjusts the protection measures according to the real-time collision prediction result to realize the optimization of the protection scheme with the change of the external environment;

[0161] S65, the system prioritizes the allocation of energy to key protection systems by linking with the energy management system, and if the vehicle is in a low energy and high energy consumption scenario, the system will automatically adjust the energy allocation scheme to ensure the normal work of the key protection;

[0162] S66, the system dynamically adjusts the protection mechanism in combination with the driver's operation when the protection mechanism is activated, and the system optimizes the protection scheme according to the driver's operation input to make the driver's operation properly respond and optimize the execution effect of the protection mechanism.

[0163] In this embodiment, S8 specifically includes:

[0164] S81, collect acceleration, braking, steering response data of the vehicle under different driving environments, and external environment data, and construct response models under different driving scenes to reflect the different influences of each scene on the vehicle behavior;

[0165] S82, through an adaptive learning algorithm, based on the collected multi-scene data, dynamically adjust the acceleration, braking and steering strategies of the vehicle, and the system optimizes the response behavior of the vehicle according to the historical operation mode of the driver and the real-time external environment changes:

[0166] R opt =f(V t ,A s ,T r ,E e );

[0167] Wherein, R opt is the optimized vehicle response, V t is the current speed of the vehicle, A s is the acceleration parameter, T r is the steering response time, and E e is the external environment factor;

[0168] S83, the system adjusts the driving strategy in different scenes according to the optimization result, in the urban road scene, the low-speed steering control is given priority to, on the highway, the system adjusts the balance of acceleration and braking, and in the mountain road, the linkage control of steering and braking is emphasized, and the system dynamically adjusts the response speed and intensity of acceleration, braking and steering according to the specific scene;

[0169] S84, the system updates the model according to the learning result by combining historical data and real-time data, and continuously updates the adaptive learning model:

[0170] M upd =M prev +α×ΔR;

[0171] Wherein, M upd is the updated model, M prev is the previous model parameter, alpha is the learning rate, and delta R is the response adjustment amount;

[0172] S85, through real-time monitoring of the driver's operation and the vehicle's running state, combined with the changes of the external environment, adjust the acceleration, braking and steering strategies of the vehicle, and according to the driver's operation frequency, road complexity and vehicle dynamics, adjust the response of the vehicle at any time;

[0173] S86, the system improves the global driving strategy in multiple scenarios through long-term driving data accumulation. Whenever the vehicle operates in a similar driving scenario, the system optimizes based on historical data and adjusts the strategy in future identical and similar scenarios. Through long-term data accumulation, the global driving strategy is optimized.

[0174] In this embodiment, the following modules are included:

[0175] UWB sensor network module: real-time collection of multi-level dynamic environment data around the vehicle and in-vehicle state information, adjustment of perception range, sensitivity and collection frequency according to the dynamic complexity of the external environment and the driving state of the vehicle;

[0176] Adaptive learning algorithm module: based on the collected data, generate a multi-level three-dimensional environment model, and dynamically adjust the sensor weights to optimize the generation and update of the model, and provide real-time environmental feedback;

[0177] Multi-factor weight evaluation module: through the multi-factor weight model of time priority, space priority, and speed priority, predict potential risks, dynamically adjust weight parameters, and generate a priority list containing real-time risk events;

[0178] Behavior prediction and path adjustment module: according to the dynamic path and behavior intention of external targets, combined with driver behavior patterns and vehicle state, adjust vehicle driving path, speed and driving mode, realize intelligent adjustment;

[0179] Passenger health monitoring module: monitor the health status of passengers through in-vehicle sensors, combine with driving path and environmental changes, automatically adjust driving mode, when abnormal health data is detected, switch to low speed or automatic driving mode;

[0180] Energy scheduling module: according to the energy demand prediction results and external environmental conditions, real-time optimization of energy distribution of key systems, in emergency priority distribution of energy to safety systems;

[0181] Collision prediction and protection module: based on multi-dimensional collision prediction model, real-time evaluation of relative position, speed, acceleration of vehicle and obstacles and external targets, activation of step-by-step protection mechanism to cope with different collision risk levels;

[0182] Driving assistance optimization module: based on multi-scenario adaptive learning algorithm, optimize acceleration, braking and steering response strategy, through long-term data accumulation, adjust global driving strategy, optimize vehicle driving performance in different driving environments.

[0183] Example 1:

[0184] In order to verify the feasibility of the application in implementation, the application is applied to the morning peak period of Sanhuan Road section in Beijing, the time is October 1, 2024, the weather is haze with light rain, the road surface is wet, the traffic flow is dense, the vehicle is in a complex urban traffic environment, surrounded by a large number of pedestrians, cyclists, cars and various traffic signals, and the weather conditions have an adverse effect on visibility, the vehicle sensor system in the prior art often cannot effectively perceive the dynamic environment and predict potential dangers under such conditions, resulting in insufficient safety, and the traditional camera and radar system is difficult to quickly and accurately identify and process multiple dynamic changes under low visibility and high traffic density, in complex traffic conditions, the burden on the driver is increased, and the system is difficult to respond to various emergencies in time, in order to solve these problems, the multifunctional integrated vehicle safety control system based on UWB of the application provides a complete solution, which significantly improves safety and response speed.

[0185] In this scenario, the UWB sensor network of the application is used to collect multi-level dynamic environment data around the vehicle in real time, compared with the traditional camera and radar, the UWB sensor has higher precision and penetration, and can accurately capture the relative position, speed and motion direction of pedestrians, cyclists and other vehicles in foggy weather and low visibility, during driving, the UWB sensor perceives that a cyclist suddenly changes lanes at a distance of 50 meters in front of the vehicle, the system immediately analyzes the motion trajectory based on the data provided by the sensor, and combines the current driving state of the vehicle to quickly evaluate the potential collision risk through a multi-factor weight model, the system predicts that the behavior of the cyclist will cause the vehicle to approach within 2 meters within 5 seconds, based on this risk assessment, the system decides to adjust the path, and adjusts the driving trajectory of the vehicle to the right side, avoiding the possible collision, the whole adjustment process only takes 500 milliseconds, which is about 40% faster than the traditional system, ensuring the safe operation of the vehicle in a high-density traffic environment.

[0186] When the vehicle passes through a busy intersection, the system detects that a vehicle coming from the side is accelerating to approach, through a multi-dimensional collision prediction model, the system analyzes the relative position, speed and acceleration of the two vehicles, and predicts that the risk of collision is 72%, the system immediately activates the collision protection mechanism, first sends an audible and light warning signal to the driver, prompting to slow down, when the two vehicles are 6 meters apart, the system automatically starts emergency braking, avoiding the potential side collision, the response time of the whole collision warning and protection mechanism is 700 milliseconds, which is reduced by 40% compared with the traditional collision warning system, ensuring the safety of the vehicle in emergency situations.

[0187] In terms of energy management, the application uses an adaptive energy scheduling model to dynamically allocate energy according to the real-time driving state of the vehicle and external environmental conditions. In congested traffic, due to frequent start-stop operations, the system prioritizes energy allocation to the braking and steering systems, ensuring that critical systems can operate normally under high load conditions. In the event of sudden weather changes, the system detects increased rainfall and immediately allocates more energy to the air conditioning and wiper systems, improving the driver's visibility and comfort. When detecting excessive energy consumption, the system further optimizes energy usage, ensuring the operation of critical safety systems while reducing unnecessary energy consumption. This energy scheduling strategy improves the overall energy efficiency of the vehicle, avoids energy waste, and ensures that critical systems such as braking and steering always have sufficient energy supply, especially in emergency situations.

[0188] The system also monitors the health status of passengers in real-time through in-vehicle sensors, including heart rate, respiratory rate, and body temperature data. When detecting an abnormal increase in the passenger's heart rate, the system immediately analyzes the current driving path, external environment, and weather conditions, and automatically switches the vehicle's driving mode to low speed to reduce the impact of emergency operations on the passenger's physical condition. In this way, the system effectively ensures the safety of passengers in complex traffic environments and responds quickly to health abnormalities. This passenger health monitoring and driving mode linkage design not only enhances the intelligence of the system but also adds an important protective measure for vehicle safety management.

[0189] Through real-time data collection by UWB sensors, the system can accurately capture dynamic information of obstacles and external targets within a 50-meter range, with a perception accuracy improvement of over 30%. The system completes path adjustment within 500 milliseconds, a 40% improvement over traditional systems. In terms of collision prediction, the system's reaction time in a side collision scenario is 700 milliseconds, successfully avoiding up to 72% of collision risks. The energy management system prioritizes allocating 35% of energy to braking and steering systems in traffic congestion and adverse weather conditions, while increasing energy to air conditioning and wipers by 20%, ensuring efficient vehicle operation. The health monitoring system switches driving mode in less than 2 seconds after detecting abnormal passenger heart rate, effectively ensuring passenger safety.

[0190] Table 1 Vehicle safety control system related data table

[0191]

[0192] Through the table data, it is known that the vehicle realizes the environment perception within 50 meters through the UWB sensor network, the perception accuracy reaches ±10 centimeters, the perception ability in complex environment is significantly improved, when encountering sudden situations such as lane changing of pedestrians or cyclists, the system can complete path adjustment within 500 milliseconds, thereby avoiding potential collision risks, through the multi-dimensional collision prediction model, the system can predict 72% of the collision risks under real-time perception, and make a protective response within 700 milliseconds, start deceleration or emergency braking, successfully avoid side collision, in terms of energy management, the system preferentially allocates 35% of the energy to the braking and steering systems when traffic congestion occurs, ensures the stable operation of the key systems, at the same time, when the weather deteriorates, 20% of the energy is allocated to the air conditioner and wiper, the driving comfort and safety are improved, when detecting abnormal passenger health, the system can switch to low-speed or automatic driving mode within 2 seconds, ensure the safety of passengers, and dynamically adjust the driving mode combined with the external environment, the advantages of intelligent health monitoring and driving linkage are shown.

[0193] In summary, through the sensor network, adaptive learning algorithm, intelligent energy scheduling and protection mechanism and other technical innovations, the safety control and energy efficiency optimization of the vehicle in complex environment are successfully realized, the overall performance of the system is significantly improved, and the safety and intelligent level that cannot be achieved by the prior art are achieved.

[0194] The above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited to this, any skilled person in the art can make equivalent replacement or change according to the technical scheme and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A multi-functional integrated vehicle safety control method based on UWB, characterized in that, Comprise the following steps: S1, through the UWB sensor network arranged around the vehicle, real-time collection of multi-level dynamic environment data around the vehicle and in-vehicle state information, the sensor automatically adjusts the sensing range, sensitivity and frequency according to the dynamic complexity of the external environment and the current driving state of the vehicle, and generates a multi-level three-dimensional environment model through an adaptive learning algorithm; S2, based on the generated multi-level three-dimensional environment model, a multi-factor weight model of time priority, space priority, speed priority and dynamic behavior prediction is introduced to predict and evaluate potential risk events, and the model dynamically adjusts the risk weight parameters according to the changes of the environment and the vehicle state, and automatically generates a priority list containing real-time risk events; S3, based on the generated priority list, the system obtains the dynamic path and behavior intention data of the external target through UWB communication with external devices, the system combines the path adjustment of the vehicle with the behavior prediction result of the external target, and fuses with the behavior mode database of the driver to intelligently adjust the driving path, speed and driving mode; S4, the system actively adjusts the driving mode of the vehicle based on the real-time monitoring of the passenger health state data and in combination with the path adjustment and external environment change, and when the passenger health data is abnormal, the system preferentially switches to low speed and automatic driving mode according to the environmental complexity and the results of the vehicle path dynamic adjustment; S5, in combination with the generated three-dimensional environment model and the path adjustment data, the energy demand is predicted according to the current environment, driving state and future driving path, the adaptive energy scheduling model is used to optimize the energy distribution of the vehicle in real time, and the system pre-plans the energy consumption priority in different scenes according to external factors, and preferentially guarantees the energy supply of key equipment in emergency; S6, the system uses a multi-dimensional collision prediction model to predict the complex collision scenarios based on the generated three-dimensional environment model and path adjustment information, and the system predicts the potential collision type by analyzing the relative position, speed, acceleration and deceleration trend and direction change of the vehicle, obstacles and external targets, and gradually activates the adaptive multi-stage protection mechanism according to the severity and probability of the collision prediction; S7, the system analyzes the current state of the driver and the external environment condition through the behavior mode fusion mechanism based on the driver's behavior mode library, automatically matches the behavior mode of the driver and adjusts the in-vehicle settings when the driver approaches the vehicle, and optimizes the vehicle start process and provides a seamless entry experience according to the current energy state of the vehicle and the behavior mode of the driver; S8, through the multi-scene adaptive learning algorithm, the driving assistance functions in each scene are optimized and adjusted, the acceleration, steering and braking response are optimized in combination with the external environment change, the driver operation mode and the vehicle performance, and in similar environment, the system further optimizes the driving strategy based on the previous optimization results to realize adaptive multi-scene learning of the vehicle.

2. The multi-functional integrated vehicle safety control method based on UWB according to claim 1, characterized in that, The S1 specifically comprises: S11, real-time collection of multi-level dynamic environment data and in-vehicle state information around the vehicle through multiple UWB sensors arranged around the vehicle, the environment data including the spatial position, relative speed, terrain change, weather condition and traffic flow density of obstacles, and the data being classified into static environment data and dynamic environment data; S12, real-time evaluation of environment complexity based on the collected static and dynamic environment data, real-time adjustment of the sampling frequency and accuracy of the sensors to adapt to the complex changes of the environment; ； wherein, represents the environmental complexity, represents the traffic flow density, represents the influence of weather conditions, is the current speed of the vehicle, represents the relative movement change of obstacles, and is an adjustment factor, is a stabilizing term; S13, dynamic adjustment of the data weight of each sensor by the system based on the evaluation result of the environment complexity; ； wherein, is a weight for sensor data, is an adaptive learning rate, is a result of weight adjustment according to environmental complexity, denotes a weight factor for obstacle changes, is a balancing factor; S14, generation of a multi-level three-dimensional environment model by the system according to the weighted sensor data after the adjustment of the sensor weight, the model including the position of dynamic obstacles, the slope change of terrain, the traffic flow density and its change trend, and continuous updating of the three-dimensional environment model according to the real-time data and the change of the sensor weight; ； wherein, is a three-dimensional environment model, represents a data packet collected by a sensor at a time instance, is a sensor weight for the instance. 3.The UWB-based multifunctional integrated vehicle safety control method according to claim 1, characterized in that, The S2 specifically includes: S21, acquisition of factor information related to time, space and speed by the system based on the generated multi-level three-dimensional environment model; S22, introduction of a multi-factor weight model prioritizing time, space and speed for weighted calculation of the time, space and speed factors: ； wherein, is a risk weight, is a time factor, is a space factor, is a speed factor, , , is a weight coefficient, is a time change amount; S23, introduction of a dynamic behavior prediction factor by the system based on the weight model for predicting the behavior path of external targets, prediction of the future behavior of the external targets by using the motion information of the external targets and combining historical data and real-time data: ； in, Represents the predicted value of dynamic behavior. For external targets in time The movement trajectory, The vehicle's current speed. The distance between the target and the vehicle; S24, the system generates a priority list of risk events according to time, space, speed factor and dynamic behavior prediction value , generates a priority list of risk events, sorts through the priority list, high-priority events are processed first, and low-priority events enter the waiting queue. 4.The UWB-based multifunctional integrated vehicle safety control method according to claim 1, characterized in that, The S3 specifically includes: S31, acquisition of dynamic path and behavior intention data of external targets by the system through UWB communication with external devices according to the generated priority list, the dynamic path and behavior intention data of external targets including the relative position, speed, direction and behavior prediction of the targets; S32, path adjustment combined with the system according to the dynamic behavior prediction value of external targets and the driving state of the vehicle: ； wherein, is a path adjustment amount, is a current speed of the vehicle, is a relative distance to an external target, is a dynamic behavior prediction value, is an adjustment coefficient, is a response time, is a prediction weight, is a stabilization term; S33, combination of the behavior prediction of external targets and the historical behavior pattern of the driver by the system to optimize the path, speed and driving mode of the vehicle, and intelligent adjustment according to the behavior pattern of the driver and the change of the current external environment; S34, adjusting based on path And intelligent adjustment, dynamic adjustment of the driving speed and driving mode of the vehicle, the adjustment of speed and path linkage, so that the vehicle can maintain the safety and stability of driving in complex external environment, through real-time analysis of external target data and driver's behavior mode, select the optimal driving mode and speed adjustment scheme. 5.The UWB-based multifunctional integrated vehicle safety control method according to claim 1, characterized in that, The S4 specifically includes: S41, real-time monitoring of the health status of passengers by the sensors in the vehicle, including heart rate, breathing rate, body temperature and posture change, real-time analysis according to the health status data and identification of any abnormal condition; S42, combination of the acquired health status data with the current driving path and external environment condition, and decision of whether to adjust the driving mode by the system through comprehensive consideration of the current driving path and weather condition if the health status of passengers is abnormal; S43、system adjusts the route according to the passenger health data , determine whether to adjust the speed and path of the vehicle: ； wherein, is a health status-based adjustment amount, denotes a change in the passenger health status, is a path adjustment amount, is an adjustment coefficient, is a time interval for health data collection, is a stabilization term; S44, automatic switching to low-speed mode or automatic driving mode by the system according to the road complexity and weather condition when detecting abnormal health status of passengers, and the switching of the driving mode according to the emergency degree of the health status of passengers and the risk evaluation result of the external environment. S45, continuously update the three-dimensional environment model by real-time monitoring of passenger health status and combining UWB communication of external devices, based on the updated environment model, the system further optimizes the driving mode and path adjustment to match the passenger health status with the driving environment, and provides a real-time response driving scheme. 6.The UWB-based multifunctional integrated vehicle safety control method according to claim 1, characterized in that, The S5 specifically includes: S51, based on the current driving path, external environment and vehicle operating state, predict energy demand in different driving scenarios, combine real-time road conditions, driving speed and vehicle load, match energy demand with actual driving scenarios, and the system adjusts energy prediction value in real time according to changing environmental conditions to optimize energy utilization; S52, dynamically adjust energy demand by real-time monitoring of external environment data, after combining external factors and energy demand, adjust energy distribution scheme according to the complexity of different driving scenarios, maximize vehicle energy efficiency and driving performance; S53, the system dynamically optimizes the energy distribution of the vehicle using an adaptive energy scheduling model based on energy demand prediction results and external environmental conditions, adjusts the energy distribution of each key system in real time, and optimizes the energy distribution strategy through the built-in self-learning algorithm according to the driving habits and historical driving data of the driver; S54, when detecting abnormal passenger health status, severe weather or traffic accident emergency, the system preferentially allocates energy to key safety systems, in emergency situations, the system will adjust the energy priority according to the risk level of the emergency event, so that the key system can still maintain normal work under energy shortage; S55, the system can further optimize energy demand according to real-time feedback of external data in extreme environment, in extreme low temperature environment, the system preferentially provides energy for key systems, and adjusts energy consumption according to real-time feedback to adapt to the special scene of low temperature and high energy consumption; S56, the system continuously monitors the energy consumption data of the vehicle and external feedback information, and optimizes the energy scheduling strategy through real-time adaptive learning algorithm. 7.The UWB-based multifunctional integrated vehicle safety control method according to claim 1, characterized in that, The S6 specifically includes: S61, by collecting multi-dimensional information such as speed, acceleration, relative position and direction change of external targets, the system constructs a collision prediction model to calculate potential collision scenarios, and the system updates the collision prediction in real time according to multi-dimensional information: ； wherein, is a collision prediction value, denotes a current speed of the vehicle, is a current acceleration, is a relative distance of the external obstacle, is a change in direction of the vehicle and the obstacle; S62, based on the collision prediction result, the system calculates the potential collision risk level, and the system adjusts the risk level according to the current driving state of the vehicle and external environmental factors: ； wherein, is a collision risk level, is an environmental impact factor, is a vehicle state impact factor, is a stability coefficient; S63, when the system detects a higher risk collision, the system gradually activates the protection mechanism according to the risk level, including early warning and deceleration prompt, activates higher level protection system, realizes sequential activation of each protection measure, and protects the safety of the vehicle and passengers; S64, after activating the protection mechanism, the system continuously monitors the speed, acceleration of the vehicle and the dynamic information of external obstacles, and adjusts the protection measures according to the real-time collision prediction result, realizes the optimization of protection scheme with the change of external environment; S65, the system allocates energy to the key protection system in priority through linkage with the energy management system, and if the vehicle is in an energy shortage and high energy consumption scenario, the system will automatically adjust the energy allocation scheme to enable the key protection to work normally; S66, when the protection mechanism is activated, the system dynamically adjusts in combination with the driver's operation, and the system optimizes the protection scheme according to the driver's operation input to make the driver's operation properly respond and optimize the execution effect of the protection mechanism. 8.The UWB-based multifunctional integrated vehicle safety control method according to claim 1, characterized in that, The S8 specifically includes: S81, collect the acceleration, braking, and steering response data of the vehicle under different driving environments, as well as external environment data, and construct a response model under different driving scenarios to reflect the different influences of each scenario on the vehicle behavior; S82, through an adaptive learning algorithm, based on the collected multi-scenario data, dynamically adjust the acceleration, braking, and steering strategies of the vehicle, and the system optimizes the response behavior of the vehicle according to the historical operation mode of the driver and the real-time external environment changes: ； wherein, is an optimized vehicle response, is a current speed of the vehicle, is an acceleration parameter, is a steering response time, is an external environmental factor; S83, the system adjusts the driving strategy in different scenarios according to the optimization results, prioritizes low-speed steering control in urban road scenarios, adjusts the balance of acceleration and braking on highways, and focuses on the linkage control of steering and braking on mountain roads, and the system dynamically adjusts the response speed and intensity of acceleration, braking, and steering according to the specific scenario; S84, the system updates the model according to the learning results by combining historical data with real-time data, and continuously updates the adaptive learning model: ； wherein, is the updated model, is the previous model parameter, is the learning rate, is the response adjustment amount; S85, through real-time monitoring of the driver's operation and the vehicle's running state, and in combination with the changes in the external environment, adjust the acceleration, braking, and steering strategies of the vehicle, and adjust the vehicle's response at any time according to the driver's operation frequency, road complexity, and vehicle dynamics; S86, the system improves the global driving strategy in multiple scenarios through long-term accumulation of driving data, optimizes based on historical data whenever the vehicle operates in similar driving scenarios, and adjusts the strategy in future identical and similar scenarios, and optimizes the global driving strategy through long-term data accumulation.

9. A multi-functional integrated vehicle safety control system based on UWB, characterized in that, The following modules are included: UWB sensor network module: real-time collection of multi-level dynamic environment data around the vehicle and in-vehicle state information, adjustment of the perception range, sensitivity, and collection frequency according to the dynamic complexity of the external environment and the driving state of the vehicle; Adaptive learning algorithm module: based on the collected data, generate a multi-level three-dimensional environment model, and dynamically adjust the sensor weights to optimize the generation and update of the model, and provide real-time environmental feedback; Multi-factor weight evaluation module: through the multi-factor weight model of time priority, space priority, and speed priority, predict potential risks, dynamically adjust weight parameters, and generate a priority list containing real-time risk events; Behavior prediction and path adjustment module: according to the dynamic path and behavior intention of external targets, in combination with the driver's behavior mode and the vehicle state, adjust the vehicle's driving path, speed, and driving mode to realize intelligent adjustment; Passenger health monitoring module: monitor the health status of passengers through in-vehicle sensors, automatically adjust the driving mode in combination with the driving path and environmental changes, and switch to low-speed or automatic driving mode when abnormal health data is detected; Energy scheduling module: Real-time optimization of energy allocation for key systems based on energy demand prediction results and external environmental conditions, prioritizing energy allocation to safety systems in emergency situations; Collision prediction and protection module: Based on a multi-dimensional collision prediction model, real-time evaluation of the relative position, speed, and acceleration of the vehicle and obstacles and external targets, and activation of a step-by-step protection mechanism to address different collision risk levels; Driving assistance optimization module: Based on multi-scenario adaptive learning algorithms, optimize acceleration, braking, and steering response strategies, adjust global driving strategies through long-term data accumulation, and optimize vehicle driving performance in different driving environments.

Images