Active seat belt control method, device, vehicle, storage medium and program product

Abstract

The present application relates to the technical field of safety belt control, and discloses an active safety belt control method, device, vehicle, storage medium and program product, wherein the present application obtains global perception information based on monitoring information sent by multiple control domains, guarantees the richness, comprehensiveness and diversity of information, then extracts local information from the global perception information according to different preset scenes, obtains target perception information required by each preset scene, improves data processing efficiency, then performs fuzzy reasoning on the target perception information, quickly obtains the risk level and active safety belt control instruction corresponding to each preset scene, and finally arbitrates the risk level and active safety belt control instruction corresponding to different preset scenes, quickly and accurately determines the target active safety belt control instruction required finally, comprehensively considers the rich perception information, guarantees the data processing efficiency, ensures the driving safety of users, and improves the user experience.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of safety belt control, in particular to an active safety belt control method and device, a vehicle, a storage medium and a program product. BACKGROUND

[0002] In recent years, active adjustment type safety belts with electric motor driving capability have been applied to in-vehicle occupant restraint systems. The active safety belt can adjust the restraint force according to the driving state of the vehicle, and provide restraint protection functions for passengers in different driving scenarios. During vehicle driving, the motion state may change rapidly and dramatically, and at this time, the occupant restraint protection function needs to be determined according to the road environment, the state of the vehicle and the occupant, and the driving scenario.

[0003] The traditional force limiting safety belt and the early warning safety belt cannot respond differently to different driving scenarios, and can only protect the passengers from violent dislocation when a collision occurs. Therefore, it cannot meet the individualized protection function of other road driving scenarios, and cannot be compatible with diversified passenger driving states and rich intelligent driving scenarios. In addition, the perception parameters and decision-making models used to identify specific driving conditions are very difficult to unify, and the number of perception parameters is extremely large, the processing process is cumbersome, and there is a risk of long calculation period and large error in the safety belt decision-making instruction calculation after determining the specific driving condition, which not only affects the user's driving experience, but also is difficult to guarantee the driving safety of the user.

[0004] Therefore, how to quickly and accurately determine the restraint effect and execution instruction of the active safety belt is a key problem that needs to be solved at present. SUMMARY

[0005] Therefore, the present application provides an active safety belt control method, device, vehicle, storage medium and program product to solve the problem that the related art cannot quickly and accurately determine the execution instruction of the active safety belt.

[0006] In a first aspect, the present application provides an active safety belt control method, which comprises:

[0007] receiving different monitoring information sent by multiple control domains, and obtaining global perception information based on the monitoring information;

[0008] extracting local information from the global perception information according to different preset scenarios to obtain target perception information corresponding to each preset scenario;

[0009] performing fuzzy reasoning on the target perception information to obtain a risk level corresponding to each preset scenario, and generating an active safety belt control instruction corresponding to each preset scenario;

[0010] Based on the risk level and the active safety belt control instruction, a target active safety belt control instruction is determined, and the target active safety belt control instruction is executed.

[0011] Therefore, the global perception information is obtained based on the monitoring information sent by the plurality of control domains, the information is rich, comprehensive and diversified, then the local information extraction is performed on the global perception information according to different preset scenes, the target perception information required by each preset scene is obtained, the data processing efficiency is improved, then the fuzzy inference is performed on the target perception information, the risk level and the active safety belt control instruction corresponding to each preset scene are quickly obtained, finally, the risk level and the active safety belt control instruction corresponding to different preset scenes are arbitrated, the target active safety belt control instruction required finally is quickly and accurately determined, the rich perception information is comprehensively considered, the data processing efficiency is ensured, the driving safety of the user is ensured, and the user experience is improved.

[0012] In an optional implementation, the fuzzy inference is performed on the target perception information to obtain the risk level corresponding to each preset scene, including:

[0013] The target perception information is fuzzified to obtain fuzzy perception information;

[0014] According to the preset fuzzy control rule and the fuzzy perception information corresponding to different preset scenes respectively, a fuzzy output set used for representing the risk level corresponding to different preset scenes is determined;

[0015] Based on the fuzzy output set, the risk level corresponding to each preset scene is obtained.

[0016] Therefore, the risk level corresponding to different preset scenes is determined by fuzzifying, fuzzy inference calculation and defuzzification processing on the target perception information, so as to measure different types of potential dangers in the vehicle driving scene.

[0017] In an optional implementation, based on the fuzzy output set, the risk level corresponding to each preset scene is obtained, including:

[0018] For each preset scene, the plurality of fuzzy outputs in the fuzzy output set corresponding to the current preset scene are respectively defuzzified to obtain the quantized output corresponding to each fuzzy output;

[0019] For each preset scene, the plurality of quantized outputs corresponding to the current preset scene are weighted and summed to obtain the risk level corresponding to the current preset scene.

[0020] Therefore, by defuzzifying and weighting and summing the plurality of fuzzy outputs, the quantized value of the risk level corresponding to different preset scenes is determined, so as to measure different types of potential dangers in the vehicle driving scene, and subsequent judgment and calculation are performed according to the quantized value of the risk level.

[0021] In an optional implementation, the target active seatbelt control instruction is determined based on the risk level and the active seatbelt control instruction, including:

[0022] A target preset scene with the maximum risk level is determined based on the risk level corresponding to each preset scene.

[0023] The target active seatbelt control instruction is determined according to the active seatbelt control instruction corresponding to the target preset scene.

[0024] Thus, by comparing the risk levels corresponding to different preset scenes respectively, the scene with greater potential risk in the current driving scene of the vehicle is determined, and the target active seatbelt control instruction to be finally executed is determined.

[0025] In an optional implementation, before the target active seatbelt control instruction is executed, the method further includes:

[0026] State information of a seatbelt hardware execution mechanism is obtained, and the state information includes at least one of network communication state, motor power-on state, motor start-up and running state, seatbelt pre-execution action signal level, seatbelt pre-execution action time length, seatbelt execution action abnormal information, remaining available number of seatbelt execution action signals, and degradation execution information of seatbelt execution action.

[0027] The state of the seatbelt hardware execution mechanism is evaluated based on the state information, and when the state evaluation is passed, the step of executing the target active seatbelt control instruction is jumped to.

[0028] Thus, by evaluating the state of the seatbelt hardware execution mechanism, the safety, accuracy and rationality of instruction execution are improved.

[0029] In an optional implementation, the global perception information is obtained based on the monitoring information, including:

[0030] The monitoring information is verified for effectiveness, resampled and numerically transformed to obtain the global perception information, and the monitoring information includes at least two of vehicle motion information, chassis control information, advanced driving assistance control information, road environment information, driver and passenger state information, active driving comfort control information and collision prediction information.

[0031] Thus, by verifying the effectiveness, resampling and numerically transforming the monitoring information, unnecessary noise data or invalid data are filtered out, and the monitoring information is standardized to improve the data processing speed.

[0032] In an optional implementation, the preset scenes include at least two of a comfort promotion scene, a safety prompting scene, a dynamic support scene, an emergency protection scene, and a collision protection scene; the global perception information is subjected to local information extraction according to different preset scenes to obtain target perception information corresponding to each preset scene, including:

[0033] For the comfort promotion scene, driver and passenger state information or active driving comfort control information in the global perception information is extracted to obtain target perception information corresponding to the comfort promotion scene;

[0034] For the safety prompting scene, driver and passenger state information or advanced driving assistance control information in the global perception information is extracted to obtain target perception information corresponding to the safety prompting scene;

[0035] For the dynamic support scene, self-vehicle motion information, chassis control information, road environment information, driver and passenger state information, or collision prediction information in the global perception information is extracted to obtain target perception information corresponding to the dynamic support scene;

[0036] For the emergency protection scene, self-vehicle motion information, chassis control information, advanced driving assistance control information, road environment information, or driver and passenger state information in the global perception information is extracted to obtain target perception information corresponding to the emergency protection scene;

[0037] For the collision protection scene, driver and passenger state information or collision prediction information in the global perception information is extracted to obtain target perception information corresponding to the collision protection scene.

[0038] Thus, by extracting target perception information required by different preset scenes, the number of input signals of the corresponding scene sub-modules and the coupling degree between the input signals are reduced, and the complexity of the model is reduced.

[0039] In an optional implementation, the target perception information is subjected to fuzzy reasoning to generate active seatbelt control instructions corresponding to each preset scene, including:

[0040] The target perception information corresponding to the comfort promotion scene is subjected to fuzzy reasoning to generate a roll-back instruction;

[0041] The target perception information corresponding to the safety prompting scene is subjected to fuzzy reasoning to generate a prompting instruction;

[0042] The target perception information corresponding to the dynamic support scene is subjected to fuzzy reasoning to generate a tightening instruction;

[0043] The target perception information corresponding to the emergency protection scene is subjected to fuzzy reasoning to generate a pull-back instruction;

[0044] And / or, the pre-warning instruction is generated based on fuzzy reasoning of the target perception information corresponding to the collision protection scene.

[0045] Therefore, the complexity is reduced and the processing efficiency is improved by fuzzy processing of the target perception information corresponding to each preset scene, and the active safety belt control instruction corresponding to each preset scene can be quickly obtained.

[0046] In a second aspect, the present application provides an active safety belt control device, which comprises:

[0047] The acquisition module is configured to receive different monitoring information sent by multiple control domains, and obtain global perception information based on the monitoring information.

[0048] The first processing module is configured to extract local information from the global perception information according to different preset scenes, and obtain target perception information corresponding to each preset scene.

[0049] The second processing module is configured to perform fuzzy reasoning on the target perception information, obtain a risk level corresponding to each preset scene, and generate an active safety belt control instruction corresponding to each preset scene.

[0050] The control module is configured to determine a target active safety belt control instruction based on the risk level and the active safety belt control instruction, and execute the target active safety belt control instruction.

[0051] In a third aspect, the present application provides a vehicle, which comprises a memory and a processor, the memory and the processor are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the active safety belt control method of the first aspect or any of the corresponding embodiments thereof.

[0052] In a fourth aspect, the present application provides a computer readable storage medium, which stores computer instructions, and the computer instructions are used to make a computer execute the active safety belt control method of the first aspect or any of the corresponding embodiments thereof.

[0053] In a fifth aspect, the present application provides a computer program product, which comprises computer instructions, and the computer instructions are used to make a computer execute the active safety belt control method of the first aspect or any of the corresponding embodiments thereof.

[0054] The present application has the following beneficial effects:

[0055] The global perception information is obtained based on the monitoring information sent by multiple control domains, the rich, comprehensive and diversified information is ensured, then the local information extraction is performed on the global perception information according to different preset scenes, the target perception information required by each preset scene is obtained, the data processing efficiency is improved, then the fuzzy reasoning is performed on the target perception information, the risk level corresponding to each preset scene and the active safety belt control instruction are quickly obtained, finally, the risk level corresponding to different preset scenes and the active safety belt control instruction are arbitrated, the target active safety belt control instruction required finally is quickly and accurately determined, the rich perception information is comprehensively considered, the data processing efficiency is ensured, the driving safety of the user is ensured, and the user experience is improved. BRIEF DESCRIPTION OF DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the specific embodiments or the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0057] Figure 1 It is a flowchart of an active safety belt control method according to an embodiment of the present application;

[0058] Figure 2 It is a flowchart of another active safety belt control method according to an embodiment of the present application;

[0059] Figure 3 It is a flowchart of still another active safety belt control method according to an embodiment of the present application;

[0060] Figure 4 It is a logic diagram of an active safety belt control method according to an embodiment of the present application;

[0061] Figure 5 It is a logic diagram of a fuzzy reasoning according to an embodiment of the present application;

[0062] Figure 6 It is a logic diagram of a risk level decision according to an embodiment of the present application;

[0063] Figure 7 It is a structure block diagram of an active safety belt control device according to an embodiment of the present application;

[0064] Figure 8 It is a hardware structure diagram of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0065] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0066] In recent years, active adjustment safety belts with electric motor driving capability have been applied to in-vehicle occupant restraint systems. Active safety belts can adjust the restraint force according to the driving state of the vehicle, and provide restraint protection functions for passengers in different driving scenarios. During vehicle driving, the motion state may change rapidly and dramatically, and at this time, the occupant restraint protection function suitable for the driving scenario needs to be determined according to the road environment, the state of the vehicle and the occupant.

[0067] Nowadays, vehicle driving is changing towards the trend of human driving as the main and intelligent driving as the auxiliary. Therefore, more complex road traffic forms will inevitably be encountered, and more accurate safety belt restraint protection strategies need to be provided for the driver and passengers to cope with different vehicle driving scenarios. The traditional force-limiting safety belt and the pre-warning safety belt cannot respond differently to different driving scenarios, and can only protect the driver and passengers from violent dislocation when a collision occurs. Therefore, they cannot meet the individualized protection functions of other road driving scenarios, and cannot be compatible with diversified occupant driving states and rich intelligent driving scenarios.

[0068] For different driving conditions, their significant influencing factors are different, and it is difficult to establish a unified decision-making model. For (critical) boundary conditions, inconsistency in control decisions is likely to occur. Even if the driving conditions are classified, decision-making based on significant influencing factor thresholds is a local optimization strategy in the average level sense (group universality, time domain invariance), but it does not consider the influence of other factors and has a certain passivity.

[0069] And, due to the perception uncertainty (accuracy of state measurement of the driving environment and other road users), the prediction uncertainty (single-agent, vehicle state change and driver intention inference), the interaction uncertainty (multi-agent, corresponding changes in the environment caused by changes in the state of the ego vehicle or the opposite case) and the execution uncertainty (reference-output, error between actual output and reference input), the determination of the driving condition also has various uncertainties. Therefore, the description of the driving condition itself has a certain uncertainty, which is more likely to increase in a dangerous scenario (state changes dramatically). The uncertainty itself implies the ambiguity of the state, and various factors need to be considered to minimize the impact of uncertainty on the decision output.

[0070] In addition, the model framework that activates all classification modules for calculation at the same time and finally arbitrates according to the results of each module has higher hardware requirements; on the other hand, more dangerous conditions relative to the secondary dangerous condition, the decision-making time is relatively late, and the decision-making calculation result is more likely to jump, that is, the trigger condition of the secondary dangerous condition is met first, and then the trigger condition of the more dangerous condition is met, which is equivalent to a scene triggering twice. The continuous change of the driving state, especially the dramatic state change, is easy to cause function mis-triggering (satisfying the condition threshold) and function trigger failure (short trigger duration). When multiple conditions meet the trigger condition at the same time, decision output conflicts or frequent jumps are easy to occur, which will cause the action error of the execution mechanism, not only affecting the service life of the execution mechanism, but also reducing the driving experience of the passengers.

[0071] The patent "CN113602053A" provides a control method and control system of active safety belt based on air suspension, which synchronously controls the active safety belt of the vehicle according to different working conditions of the vehicle, but the working conditions involved are all determined based on the advanced driving assistance system (ADAS), which is highly dependent on the function list of the ADAS system. The patent "CN204340966U" provides an active safety belt device for a car that can automatically adjust according to the actual road conditions, which identifies and judges the actual road conditions through an image recognition and processing module, automatically tightens or loosens the safety belt, and reduces the risk of personal injury caused by safety accidents, but the working conditions involved are only for collision conditions, and the application scenario is narrow.

[0072] Therefore, the embodiment of the present application provides an active safety belt control method, which can quickly and accurately determine the restraint effect and execution instruction of the active safety belt, improve the driving safety guarantee of the driver and passenger, increase the driving comfort experience, and improve the driving experience.

[0073] According to the embodiment of the present application, an active seatbelt control method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in a different order.

[0074] In the present embodiment, an active seatbelt control method is provided, which can be used in a vehicle equipped with an active seatbelt, Figure 1 is a flowchart of the active seatbelt control method according to the embodiment of the present application, as shown in Figure 1 The flowchart includes the following steps:

[0075] In step S101, different monitoring information sent by multiple control domains is received, and global perception information is obtained based on the monitoring information.

[0076] Specifically, the monitoring information is verified for validity, resampled and numerically transformed to obtain the global perception information, and unnecessary noise data or invalid data is filtered out, and the monitoring information is standardized to improve the data processing speed. The monitoring information includes at least two of vehicle motion information, chassis control information, advanced driving assistance control information, road environment information, driver and passenger state information, active driving comfort control information and collision prediction information.

[0077] It should be noted that the control domain in the embodiment of the present application refers to different perception systems or control systems, such as vehicle control systems, chassis control systems, ADAS systems, road perception systems, state recognition systems, active driving comfort control and collision prediction systems, etc. The monitoring information of different control domains can ensure the comprehensive richness and diversity of the information and improve the accuracy of the control.

[0078] For example, the vehicle motion information includes but is not limited to vehicle speed, longitudinal / lateral acceleration, heading angle, yaw rate, wheel speed, etc.

[0079] For example, the chassis control information includes but is not limited to steering wheel angle and angle speed, throttle / brake pedal opening, brake master cylinder pressure and pressure rate, motor reverse torque, gear position, four-wheel speed, etc.

[0080] Exemplarily, the advanced driving assistance control information (ADAS driving assistance information) includes, but is not limited to, ADAS switch signals, Autonomous Emergency Braking (AEB) control signals, Forward Collision Warning (FCW) control signals, Electronic Stability Control (ESC) control signals, Electrical Park Brake (EPB) control signals, Adaptive Cruise Control (ACC) control signals, Anti-Lock Brake System (ABS) control signals, Electronic Brake Assist (EBA), Electronic Stability Program (ESP), and the like.

[0081] Exemplarily, the road environment information includes, but is not limited to, road conditions, obstacle information, traffic sign recognition, and the like, especially for the detected dangerous target object set.

[0082] Exemplarily, the driver state information includes, but is not limited to, whether the seat is occupied, the wearing state of the safety belt, the gender, the body shape, the attention, the physical state, the position and the like of the driver. When the driver is seated and the safety belt is worn, the gender, the age, the height, the body shape and the like of the driver are automatically recognized, the active safety belt, the intelligent seat, the steering wheel position and the like are adaptively adjusted according to the human database formulated in advance, and the heart rate, the blood oxygen and the emotion of the driver are monitored to improve the perception accuracy of the personnel and to explore the timely needs.

[0083] Exemplarily, the Active Ride Control (ACR) information includes, but is not limited to, the network communication state of the ACR system, the action execution / idle state feedback, the remaining available number of actions corresponding to different warning levels, the seat angle, the function inhibition / ignore / closure state and the like.

[0084] Exemplarily, the collision prediction information includes, but is not limited to, the collision probability, the prediction of the vehicle collision damage degree, the prediction of the passenger collision damage degree and the like, and is used for predicting the collision.

[0085] In step S102, the global perception information is extracted into local information according to different preset scenes, and target perception information corresponding to each preset scene is obtained.

[0086] Exemplarily, according to the previous investigation and test, the embodiment of the present application can cluster the driving scene into five categories, mainly including a comfort promotion scene, a safety prompting scene, a dynamic support scene, an emergency protection scene, and a collision protection scene, and a corresponding scene discriminator can be established for different preset scenes. It should be noted that the above-mentioned scene classification rules need to ensure that the coupling overlap degree between the preset scenes is small.

[0087] In some optional embodiments, for the comfort promotion scene, the driver and passenger state information or the active driving comfort control information in the global perception information is extracted to obtain the target perception information corresponding to the comfort promotion scene. Specifically, the comfort promotion scene focuses on the driving scene without any safety risk, and the corresponding control instruction is generated mainly by monitoring the seat belt state and the driver and passenger state.

[0088] In some optional embodiments, for the safety prompting scene, the driver and passenger state information or the advanced driving assistance control information in the global perception information is extracted to obtain the target perception information corresponding to the safety prompting scene. Specifically, the safety prompting scene mainly focuses on the dangerous driving scene caused by the abnormal state of the driver and passenger and the abnormal signal of the intelligent driving assistance system, and the driver is reminded to pay attention to safety by monitoring the above signal abnormalities.

[0089] In some optional embodiments, for the dynamic support scene, the self-vehicle motion information, chassis control information, road environment information, driver and passenger state information, or collision prediction information in the global perception information is extracted to obtain the target perception information corresponding to the dynamic support scene. Specifically, the dynamic support scene mainly focuses on the potential safety problem in the vehicle lateral motion caused by the driver's operation or environmental changes during driving.

[0090] In some optional embodiments, for the emergency protection scene, the self-vehicle motion information, chassis control information, advanced driving assistance control information, road environment information, or driver and passenger state information in the global perception information is extracted to obtain the target perception information corresponding to the emergency protection scene. Specifically, the emergency protection scene mainly focuses on the potential safety problem in the vehicle longitudinal motion caused by the driver's operation or environmental changes during driving.

[0091] In some optional embodiments, for the collision protection scene, the driver and passenger state information or the collision prediction information in the global perception information is extracted to obtain the target perception information corresponding to the collision protection scene. Specifically, the collision protection scene mainly focuses on the imminent or ongoing unavoidable collision, rollover, and other dangerous events of the vehicle.

[0092] Due to a large number of input signals, many signals present a nonlinear relationship, and the unified decision algorithm structure based on all input signals (global perception information) is complex and requires a huge amount of computing power. The embodiment of the present application first clusters the scene, establishes a scene discrimination model according to the focus of different preset scenes, extracts the target perception information required by different preset scenes, reduces the number of input signals of the corresponding scene sub-module and the coupling degree between the input signals, and further reduces the complexity of the model.

[0093] In step S103, fuzzy reasoning is performed on the target perception information to obtain the risk level corresponding to each preset scene, and an active safety belt control instruction corresponding to each preset scene is generated.

[0094] Specifically, according to the different focuses of the five scene discriminators, the required target perception information is respectively transmitted into the five scene discriminators. When the scene discriminator is activated, the risk level or the execution instruction level of the active safety belt is calculated based on the preset fuzzy reasoning rule.

[0095] Exemplarily, the aforementioned monitoring information includes but is not limited to vehicle motion information, chassis control information, ADAS control information, road environment information (vehicles, lanes, obstacles, etc.), driver and passenger state information, active driving comfort control information, and collision prediction information, etc. Considering the fuzzy reasoning rule, the fuzzy relationship between the vehicle driving scene risk level is constructed, so as to quickly obtain the active safety belt control instruction at the current time. When applying multi-sensor information fusion, the fuzzy set theory can be used to represent the uncertainty of each sensor information by using the membership function, and then the fuzzy transformation is used for comprehensive processing. It should be noted that the constraint degree of the active safety belt control instruction can be determined according to the size of the risk level obtained by fuzzy reasoning. The higher the risk level, the greater the constraint degree; the lower the risk level, the smaller the constraint degree.

[0096] In some optional embodiments, the target perception information corresponding to the comfort improvement scene is subjected to fuzzy reasoning to generate a roll-back instruction. Exemplarily, after the driver and passenger first wear the safety belt, according to the observed body posture, sitting posture and other information of the driver and passenger, the active safety belt slightly adjusts the safety belt pull-out amount to eliminate the gap between the driver and passenger and the seat; when the driver and passenger unlock the safety belt, the lock state changes from the locked state to the unlocked state, and the active safety belt starts to recover the webbing until it is completely stored in place. It should be noted that the roll-back degree of the roll-back instruction can be determined according to the size of the risk level obtained by fuzzy reasoning. The higher the risk level, the greater the roll-back degree; the lower the risk level, the smaller the roll-back degree.

[0097] In some optional embodiments, the target perception information corresponding to the safety reminding scenario is subjected to fuzzy inference to generate a reminding instruction. Illustratively, whether the driver is in a state of distraction, smoking, making a phone call, or being away from the seat for a long time is observed by an image device such as a driver monitoring system (DMS) or a ToF (Time of Flight) camera; a signal of intelligent driving takeover, automatic parking assistance (APA) function delivery, or the like is sent by the ADAS control domain to trigger active seat belt vibration to remind the driver to drive safely.

[0098] In some optional embodiments, the target perception information corresponding to the dynamic support scenario is subjected to fuzzy inference to generate a tightening instruction. Illustratively, in the case that the driver has a lateral displacement due to understeering / oversteering, high-speed cornering, or the like, according to road environment information such as the number of lanes, lane line crossing, obstacle collision risk, and the like, and the driver state and off-seat condition, the active seat belt provides additional restraint force or pullback force to enable the driver to maintain a safe sitting posture.

[0099] In some optional embodiments, the target perception information corresponding to the emergency protection scenario is subjected to fuzzy inference to generate a pullback instruction. Illustratively, in the case of a dangerous situation such as the driver being away from the seat longitudinally due to the driver's human error or an emergency brake sent by the AEB system, or the activation of a vehicle stability assistance system such as ABS, EBA, or ESP, the active seat belt triggers a warning function to provide additional restraint force or pullback force to the driver according to the vehicle speed to enable the driver to maintain a safe sitting posture.

[0100] In some optional embodiments, the target perception information corresponding to the collision protection scenario is subjected to fuzzy inference to generate a warning instruction. Illustratively, a high collision probability or a high collision damage prediction signal output by a collision prediction or a collision event trigger signal sent by an airbag controller triggers the active seat belt, which generally provides the highest level of pullback. Of course, the pullback force output of the active seat belt can also be adjusted by combining a seat domain occupant state monitoring system to evaluate the body shape, body posture, and off-seat condition of the driver to ensure the safety of drivers of different sizes in dangerous working conditions.

[0101] Thus, the target perception information corresponding to each preset scenario is subjected to fuzzy processing, which reduces the complexity and improves the processing efficiency, and enables the active seat belt control instructions corresponding to different preset scenarios to be obtained quickly.

[0102] Step S104, determining a target active safety belt control instruction based on the risk level and the active safety belt control instruction, and executing the target active safety belt control instruction.

[0103] Specifically, the active safety belt control instructions corresponding to different preset scenes are arbitrated, and the state signals fed back by the safety belt hardware execution mechanism are also evaluated, so that the final active safety belt action execution signal, i.e., the target active safety belt control instruction, is determined.

[0104] The active safety belt control method provided in this embodiment obtains global perception information based on the monitoring information sent by multiple control domains, ensures the richness, comprehensiveness and diversity of the information, then extracts local information from the global perception information according to different preset scenes to obtain target perception information required by each preset scene, improves the data processing efficiency, then performs fuzzy reasoning on the target perception information to quickly obtain the risk level and the active safety belt control instruction corresponding to each preset scene, and finally arbitrates the risk level and the active safety belt control instruction corresponding to different preset scenes to quickly and accurately determine the target active safety belt control instruction required finally, which comprehensively considers the rich perception information and ensures the data processing efficiency, guarantees the driving safety of users and improves the user experience.

[0105] An active safety belt control method is provided in this embodiment, which can be used for a vehicle configured with an active safety belt, Figure 2 is a flowchart of the active safety belt control method according to an embodiment of the present application, as shown in the figure, the flowchart includes the following steps: Figure 2

[0106] Step S201, receiving different monitoring information sent by multiple control domains, and obtaining global perception information based on the monitoring information. For details, see step S101 of the embodiment shown in Figure 1 herein.

[0107] Step S202, extracting local information from the global perception information according to different preset scenes to obtain target perception information corresponding to each preset scene. For details, see step S102 of the embodiment shown in Figure 1 herein.

[0108] Step S203, performing fuzzy reasoning on the target perception information to obtain a risk level corresponding to each preset scene, and generating an active safety belt control instruction corresponding to each preset scene.

[0109] Specifically, the above step S203 includes:

[0110] Step S2031, performing fuzzy processing on the target perception information to obtain fuzzy perception information.

[0111] ​Specifically, the target perception information required for different preset scenarios is fuzzified, and typical input signals are taken as examples: for numerical signals, including but not limited to vehicle speed, longitudinal / lateral acceleration, heading angle, yaw rate, wheel speed, etc., a fuzzy language variable is established according to the value and direction, and the fuzzy language variable for the value can be in the form of {very small, small, slightly small, medium, slightly large, large, very large}, and the fuzzy language variable for the direction (relative to the vehicle coordinate system) can be in the form of {left, slightly left, medium, slightly right, right}, {backward, stationary, forward}, etc.

[0112] It should be noted that the fuzzy domain corresponding to the above fuzzy language variable can be defined according to the legal value range of each input signal, for example, the fuzzy domain corresponding to the fuzzy language variable of the vehicle speed is [0 kph, 120 kph], and the fuzzy domain corresponding to the fuzzy language variable of the longitudinal acceleration can be [0, 1.1g] (where the acceleration condition can be [0, 0.4g], and the deceleration condition can be [-1.1g, 0]). For driver and passenger state monitoring signals, for example, the fuzzy domain corresponding to the fuzzy language variable of the body shape can be defined as [150 cm, 200 cm], and the fuzzy domain corresponding to the fuzzy language variable of the pose offset can be defined as [0 mm, 200 mm], and the fuzzy language variable {small, slightly small, medium, slightly large, large} is established. For Boolean logic signals, including but not limited to ADAS signal on / off, seat occupancy, seat belt wearing, driver and passenger gender, and attention, the fuzzy process is omitted, and the {0, 1} signal directly represents {yes, no}.

[0113] In addition, the membership rules involved in the fuzzification process can be Gaussian distribution rules, stage rules, triangular rules, trapezoidal rules, etc. The membership rules used by each fuzzy variable can be different according to the effect.

[0114] Step S2032, according to the preset fuzzy control rules and fuzzy perception information corresponding to different preset scenarios, respectively, determine the fuzzy output set corresponding to different preset scenarios for representing the risk level.

[0115] Specifically, the fuzzy reasoning is an output signal reasoning and calculation process based on the aforementioned fuzzified fuzzy perception information and the preset fuzzy control rules, which has the ability to simulate human fuzzy concept reasoning. This process is based on the implication relationship and reasoning rules in fuzzy logic.

[0116] Specifically, a set of fuzzy condition statements can be pre-set according to the previous active safety belt function research and expert experience to constitute the fuzzy control rule. Through the use of the fuzzy control rule, a series of logical condition comprehensive evaluations are performed on the aforementioned fuzzy perception information to obtain a quantitative language expression, i.e., the fuzzy output corresponding to the fuzzy perception information.

[0117] In the embodiment of the present application, the fuzzy output set of the risk level of each preset scene is calculated according to the aforementioned fuzzy perception information. For each preset scene, the corresponding risk level also has a fuzzy language variable {mild, general, severe}, and the corresponding domain can be [0, 10]. Due to the performance of the safety belt itself, the calibration strategy, the functional requirements, etc., the variable domain may not be able to be divided into 10 or more, and the domain here is only an example and is not limited.

[0118] Specifically, the preset fuzzy control rule can be described as follows: if A and / or B (condition), then Y = f(A, B) (result), wherein A and B are subsets of fuzzy language variables, and f(A, B) is a function of A and B. Exemplarily, a series of fuzzy control rules can be pre-prepared according to expert experience and functional requirements, and here, only the emergency braking scene is taken as an example to briefly describe the rule establishment method:

[0119] The target perception information corresponding to the fuzzy perception information of the emergency braking scene is determined, including but not limited to vehicle speed v, deceleration a, brake pedal opening degree θ, brake master cylinder pressure change rate Δp, and ABS signal, and the output signal based on the fuzzy control rule is the fuzzy output set of the risk level of the emergency braking scene or a subset of the fuzzy output set. According to the fuzzification of the aforementioned target perception information, the fuzzy control rule can be programmed, and the following brief rules do not represent the actual calibration situation and are only used as an example to demonstrate:

[0120] Table I Fuzzy control rule of emergency braking scene

[0121]

[0122] Specifically, the fuzzy control rule under different preset scenes can be established based on the fuzzy control rule establishment method of the aforementioned emergency braking scene, and Table II below is an example of the fuzzy control rule corresponding to the emergency protection scene:

[0123] Table II Fuzzy control rule of emergency protection scene

[0124]

[0125]

[0126] Specifically, the fuzzy inference calculation process of each preset scene is based on all the corresponding fuzzy perception information, and when a certain rule in the fuzzy control rule is activated, the corresponding fuzzy output is obtained, and finally a fuzzy output set including multiple fuzzy outputs is obtained. Exemplarily, the fuzzy inference system can be a Mamdani Fis fuzzy inference system or a Sugeno Fis fuzzy inference system, and the like, which is not limited here, and one of them can be selected, or the fuzzy output results obtained after parallel connection of multiple can be multiplied by weights and summed. In addition, when formulating specific fuzzy control rules, the combination of fuzzy sets can be mapped to a specific scene (for example, an emergency protection scene), the rules can be divided into several basic fuzzy rules, and the basic fuzzy rules can be expanded into complex rules through permutation and combination and the like; or the number of basic fuzzy rules in the fuzzy control rule can be reduced, and the rule determination method can be simplified, which is not limited here.

[0127] In step S2033, based on the fuzzy output set, the risk level corresponding to each preset scene is obtained.

[0128] In some optional embodiments, the above step S2033 includes:

[0129] In step a1, for each preset scene, the multiple fuzzy outputs in the fuzzy output set corresponding to the current preset scene are respectively de-fuzzified to obtain quantized outputs corresponding to the fuzzy outputs.

[0130] In step a2, for each preset scene, the multiple quantized outputs corresponding to the current preset scene are weighted and summed to obtain the risk level corresponding to the current preset scene.

[0131] Specifically, de-fuzzification is to map the fuzzy output in the fuzzy output set to a certain quantity, and fuzzy inference is essentially a logic inference with uncertainty / probability. Each activated basic fuzzy rule has a corresponding weight coefficient. The result obtained by fuzzy inference is a fuzzy output set, i.e., a fuzzy output set composed of all activated fuzzy outputs, each fuzzy output carries a corresponding weight coefficient, and the fuzzy output set is comprehensively calculated to obtain a final determined output value (scalar), i.e., the de-fuzzification of the risk level of different preset scenes.

[0132] Thus, by de-fuzzifying and weighting and summing the multiple fuzzy outputs, the quantized values of the risk levels corresponding to different preset scenes are determined, so as to measure different types of potential dangers in the vehicle driving scene, and subsequent judgment and calculation are performed according to the quantized values of the risk levels.

[0133] Exemplarily, for the fuzzy reasoning of the foregoing emergency protection scene, different activated fuzzy outputs under the basic fuzzy rules are obtained, and a fuzzy output set Y e {mild, general, severe} of the risk level of the vehicle driving scene is composed, after defuzzification, a final output of a certain scalar value Y' e [0, 10) is obtained. The method of defuzzification is not specified in the application, including but not limited to the maximum membership degree method, the area method, the average method, etc.

[0134] Thus, by fuzzy processing, fuzzy reasoning calculation and defuzzification processing on the target perception information, the risk levels corresponding to different preset scenes are determined, so as to measure different types of potential dangers in the vehicle driving scene.

[0135] Step S2034, generating the active seat belt control instruction corresponding to each preset scene.

[0136] Specifically, the active seat belt control instruction can be generated by fuzzy reasoning according to the target perception information of each preset scene in the manner of the above steps S2031 to S2033, wherein the constraint degree of the active seat belt control instruction can be set according to the risk level.

[0137] Step S204, determining the target active seat belt control instruction based on the risk level and the active seat belt control instruction, and executing the target active seat belt control instruction.

[0138] Specifically, the above step S204 includes:

[0139] Step S2041, determining the target preset scene with the maximum risk level based on the risk levels corresponding to each preset scene.

[0140] Specifically, the arbitration process of the risk levels corresponding to different preset scenes can compare the risk levels of different aspects of the vehicle driving scene, and take the active seat belt control instruction of the preset scene corresponding to the maximum risk level value as the target active seat belt control instruction to be finally executed.

[0141] Step S2042, determining the target active seat belt control instruction according to the active seat belt control instruction corresponding to the target preset scene, and executing the target active seat belt control instruction.

[0142] Specifically, the arbitration process of the risk levels corresponding to different preset scenes can compare the risk levels of different aspects of the vehicle driving scene, and take the active seat belt control instruction of the preset scene corresponding to the maximum risk level value as the target active seat belt control instruction to be finally executed.

[0143] Thus, by comparing the risk levels corresponding to different preset scenes respectively, the scene with greater potential risk in the current driving scene of the vehicle is determined, and thus the target active seat belt control instruction to be finally executed is determined.

[0144] In some optional embodiments, before executing the target active seat belt control instruction, the following steps are further performed:

[0145] Step b1, obtaining state information of the seat belt hardware execution mechanism.

[0146] Step b2, performing state evaluation on the seat belt hardware execution mechanism based on the state information, and when the state evaluation passes, jumping to the step of executing the target active seat belt control instruction.

[0147] Exemplarily, the state information includes at least one of the following: network communication state, motor power-on state (whether the seat belt execution hardware is normal), motor start running state (whether the seat belt is in a certain action execution process), safety belt pre-execution action signal level (comparing the signal level to determine whether to ignore / override the currently executing action), safety belt pre-execution action duration (whether to stop / extend / override the currently executing action), safety belt execution action abnormal information (jumping signal), remaining available times of safety belt execution action signal (life calculation of the execution action signal), and degradation execution information of safety belt execution action (degradation processing of the execution action whose usage times have been used up). When the state evaluation passes and the hardware execution mechanism allows, the target active seat belt control instruction is executed, thereby improving the safety, accuracy and rationality of instruction execution.

[0148] The active seat belt control method provided in this embodiment obtains global perception information based on the monitoring information sent by multiple control domains, ensures the richness, comprehensiveness and diversity of the information, then extracts local information from the global perception information according to different preset scenes to obtain target perception information required by each preset scene, improves the data processing efficiency, then performs fuzzification, fuzzy reasoning calculation and defuzzification processing on the target perception information, quickly obtains the risk level and the active seat belt control instruction corresponding to each preset scene, and finally takes the active seat belt control instruction corresponding to the preset scene with the greatest risk level as the target active seat belt control instruction to be finally required, which comprehensively considers the rich perception information, ensures the data processing efficiency, guarantees the driving safety of the user, and improves the user experience.

[0149] The active safety belt control method of the embodiment of the application is described in detail below in combination with a specific application example. The application example provides a precise active safety belt control method, which quickly responds to the vehicle motion state, focuses on comfort experience and appropriately reminds the driver and passengers in general driving state, focuses on driving safety and gives a warning pullback before a "real" dangerous state occurs, and can meet the safety protection of the driver and passengers in all scenarios and increase the driving experience. By considering driving safety and comfort experience at the same time, different driving conditions are classified and graded, and the control decision of the active safety belt is quickly and accurately determined.

[0150] The application example considers the uncertainty of driving condition judgment including elements such as people, vehicles and roads, and determines the control decision of the active safety belt according to the risk level of the driving condition, as shown in Figure 3 The application example considers the uncertainty of driving condition judgment including elements such as people, vehicles and roads, and determines the control decision of the active safety belt according to the risk level of the driving condition, as shown in

[0151] Step S301, factors affecting the safety belt control strategy are obtained, mainly including vehicle motion information (such as speed, acceleration, heading angle, etc.), driving environment parameters including road conditions, weather, road types, etc.), road environment information (including vehicles, pedestrians, sharp turns, broken roads, etc.), driver and passenger state information (including driving age, gender, attention, safety restraint system state, etc.) and the like.

[0152] Step S302, the signal pre-processing module receives different monitoring information sent by multiple control domains, including but not limited to vehicle motion information, active driving comfort control information, etc. In particular, collision prediction information, chassis control information, ADAS intelligent driving information, road environment information, and driver and passenger state information are received. After the signal pre-processing module verifies the effectiveness and time synchronization of the received monitoring signals, the global perception information is obtained and transmitted to the subsequent module as a signal input.

[0153] Step S303, the scene clustering and reasoning control module takes the global perception information of the signal pre-processing module as input, evaluates the risk level of the driving scene, and outputs the control decision of the active safety belt, i.e. the target active safety belt control instruction.

[0154] Specifically, according to the preliminary investigation and test, the driving scene can be divided into several categories based on the risk level, including but not limited to comfort improvement scene, safety reminder scene, dynamic support scene, emergency protection scene and collision protection scene, etc.

[0155] Step S304, the signal post-processing module sends the control decision of the active safety belt to the safety belt hardware execution mechanism, so that the safety belt hardware execution mechanism performs corresponding actions.

[0156] AsFigure 4 As shown, the scene clustering and inference control module mainly includes a scene clustering unit 401 and an inference control unit 402.

[0157] Scene clustering unit 401 is used to establish five categories of scene discriminators based on preliminary research and experiments, classifying driving scenarios. Due to the large number of input signals and the non-linear relationships between many signals, a unified decision-making algorithm based on all input signals is complex and requires enormous computational power. This invention first clusters the scenes, and then establishes scene discrimination models according to the focus of different preset scenes. By extracting the target perception information required by each preset scene, the number of input signals for the corresponding scene sub-modules and the coupling between input signals are reduced, thereby reducing the complexity of the model.

[0158] Specifically, corresponding control strategies can be formulated based on the emphasis on experience / safety in different scenarios. This invention separates driving scenario classification from risk level grading. First, the scenario classification is judged, which is equivalent to filtering all input signals. Then, the corresponding risk level grading submodule is activated based on the scenario classification result to make a decision, thereby effectively reducing the amount of computation and the number of functions that are falsely triggered.

[0159] The inference control unit 402 is used to determine the active seatbelt control command based on preset fuzzy control rules. That is, it uses the aforementioned monitoring information, including but not limited to vehicle motion information, chassis control information, ADAS control information, road environment information (vehicle, lane, obstacles, etc.), occupant status information, active driving comfort control information, and collision prediction information, such as... Figure 5 As shown, considering the fuzzy inference rules, a fuzzy relationship is constructed between the rules and the risk level of the vehicle driving scenario, thereby quickly obtaining the active seat belt control command at the current moment. When applying multi-sensor information fusion, fuzzy set theory can use membership functions to represent the uncertainty of each sensor's information, and then use fuzzy transform for comprehensive processing.

[0160] like Figure 6 As shown, the signal post-processing module prioritizes and arbitrates multiple control decisions (active seatbelt control commands) from the scene clustering and inference control module. Simultaneously, it evaluates the status information fed back by the seatbelt hardware actuator to determine the final target active seatbelt control command to be executed. The arbitration of multiple decision signals compares and comprehensively evaluates the risk levels corresponding to active seatbelt control commands in multiple different preset scenarios to determine the final execution signal. Furthermore, it utilizes safety information provided by the seatbelt hardware actuator to improve the safety, accuracy, and rationality of command execution.

[0161] The application comprehensively considers all safety-related influencing factors of people, vehicles and roads, provides a full-scene active safety belt control method for vehicle driving, focuses on comfort experience and appropriately reminds the driver and passengers in a low risk level state, focuses on driving safety and performs safety belt warning and pulling back in a high safety risk level state, meets full-function and full-scene safety protection of the driver and passengers, and increases driving experience.

[0162] The embodiment of the application considers the uncertainty of driving condition judgment, reduces the high-precision dependency of the determination of the certainty threshold, and is not limited to the function list provided by the ADAS system. In addition, the state signal of the driver and passengers is considered, the driver and passengers are identified whether there is a longitudinal / lateral off-site condition through the occupant state monitoring system in the cabin, so as to determine the control decision output of the active safety belt.

[0163] In the embodiment of the application, the reasoning calculation is not specific to each driving condition, and the driving condition of the vehicle is not actively identified, but when the specific fuzzy control rule is formulated, the rule is divided into a plurality of basic fuzzy rules, and the basic fuzzy rules are expanded into complex rules through permutation and combination, so that the determination of the fuzzy control rule is simplified, and the number of rules is reduced. According to the finite sub-covering theory of open sets, all complex rules can be exhausted to meet the corresponding scene judgment requirements, so as to complete the formulation of the fuzzy control rule; and the fuzzy control rule is convenient for subsequent modification and addition of new basic rules.

[0164] In the embodiment, an active safety belt control device is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and has been described above and will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware or a combination of software and hardware is also possible and is contemplated.

[0165] The embodiment provides an active safety belt control device, as shown in Figure 7 , comprising:

[0166] The acquisition module 701 is configured to receive different monitoring information sent by a plurality of control domains, and obtain global perception information based on the monitoring information.

[0167] The first processing module 702 is configured to extract local information from the global perception information according to different preset scenes, and obtain target perception information corresponding to each preset scene.

[0168] The second processing module 703 is configured to perform fuzzy reasoning on the target perception information, obtain a risk level corresponding to each preset scene, and generate an active safety belt control instruction corresponding to each preset scene.

[0169] The control module 704 is configured to determine a target active seatbelt control instruction based on the risk level and the active seatbelt control instruction, and execute the target active seatbelt control instruction.

[0170] In some optional embodiments, the acquisition module 701 is further configured to:

[0171] The monitoring information includes at least two of the following: vehicle motion information, chassis control information, advanced driving assistance control information, road environment information, passenger state information, active driving comfort control information, and collision prediction information.

[0172] In some optional embodiments, the preset scenarios include at least two of the following: a comfort improvement scenario, a safety reminder scenario, a dynamic support scenario, an emergency protection scenario, and a collision protection scenario; and the first processing module 702 is further configured to:

[0173] For the comfort improvement scenario, the passenger state information or the active driving comfort control information in the global perception information is extracted to obtain target perception information corresponding to the comfort improvement scenario.

[0174] For the safety reminder scenario, the passenger state information or the advanced driving assistance control information in the global perception information is extracted to obtain target perception information corresponding to the safety reminder scenario.

[0175] For the dynamic support scenario, the vehicle motion information, the chassis control information, the road environment information, the passenger state information, or the collision prediction information in the global perception information is extracted to obtain target perception information corresponding to the dynamic support scenario.

[0176] For the emergency protection scenario, the vehicle motion information, the chassis control information, the advanced driving assistance control information, the road environment information, or the passenger state information in the global perception information is extracted to obtain target perception information corresponding to the emergency protection scenario.

[0177] For the collision protection scenario, the passenger state information or the collision prediction information in the global perception information is extracted to obtain target perception information corresponding to the collision protection scenario.

[0178] In some optional embodiments, the second processing module 703 is further configured to:

[0179] Perform fuzzy reasoning based on the target perception information corresponding to the comfort improvement scenario to generate a roll-back instruction.

[0180] Perform fuzzy reasoning based on the target perception information corresponding to the safety reminder scenario to generate a reminder instruction.

[0181] and / or, performing fuzzy inference based on the target perception information corresponding to the dynamic support scenario to generate a tightening instruction;

[0182] and / or, performing fuzzy inference based on the target perception information corresponding to the emergency protection scenario to generate a retraction instruction;

[0183] and / or, performing fuzzy inference based on the target perception information corresponding to the collision protection scenario to generate a warning instruction.

[0184] In some optional embodiments, the second processing module 703 is further configured to:

[0185] fuzzifying the target perception information to obtain fuzzy perception information;

[0186] determining, according to the preset fuzzy control rules corresponding to different preset scenarios respectively and the fuzzy perception information, a fuzzy output set corresponding to each preset scenario and used for representing a risk level;

[0187] obtaining, based on the fuzzy output set, the risk level corresponding to each preset scenario.

[0188] In some optional embodiments, the second processing module 703 is further configured to:

[0189] performing, for each preset scenario, de-fuzzification processing on each fuzzy output in the fuzzy output set corresponding to the current preset scenario to obtain a quantized output corresponding to each fuzzy output;

[0190] performing, for each preset scenario, weighted summation on the quantized outputs corresponding to the current preset scenario to obtain the risk level corresponding to the current preset scenario.

[0191] In some optional embodiments, the control module 704 is further configured to:

[0192] determining a target preset scenario with the largest risk level based on the risk levels corresponding to the preset scenarios;

[0193] determining a target active seat belt control instruction according to the active seat belt control instruction corresponding to the target preset scenario.

[0194] In some optional embodiments, the control module 704 is further configured to:

[0195] obtaining state information of a seat belt hardware execution mechanism; wherein the state information includes at least one of network communication state, motor power-on state, motor start-up and running state, seat belt pre-execution action signal level, seat belt pre-execution action time length, seat belt execution action abnormal information, remaining available number of seat belt execution action signals, and degradation execution information of seat belt execution action.

[0196] The status of the seat belt hardware actuator is evaluated based on the status information. Once the status evaluation is passed, the process jumps to the step of executing the target active seat belt control command.

[0197] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0198] In this embodiment, the active seatbelt control device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0199] This invention also provides a vehicle having the above-described features. Figure 7 The active seatbelt control device shown.

[0200] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of a vehicle provided in an optional embodiment of the present invention, such as... Figure 8 As shown, the vehicle includes one or more processors 10, memory 20, and interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces. The various components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the vehicle, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple vehicles can be connected, with each device providing some of the necessary operations (e.g., as a server array, a set of blade servers, or a multiprocessor system). Figure 8 Take a processor 10 as an example.

[0201] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GPA), or any combination thereof.

[0202] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0203] The memory 20 can include a program storage area and a data storage area. The program storage area can store an operating system, application programs required for at least one function, and the like. The data storage area can store data created based on use of the vehicle, and the like. In addition, the memory 20 can include a high-speed random access memory, and can also include a non-transitory memory such as at least one of a magnetic disk storage device, a flash memory device, or other non-transitory solid state memory device. In some alternative embodiments, the memory 20 can optionally include memory that is remotely located with respect to the processor 10, and these remotely located memories can be connected to the vehicle through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0204] The memory 20 can include a volatile memory such as a random access memory, and can also include a non-volatile memory such as a flash memory, a hard disk, or a solid state disk. The memory 20 can also include a combination of the above-mentioned types of memory.

[0205] The vehicle also includes a communication interface 30 for communication of the vehicle with other devices or communication networks.

[0206] The embodiments of the present application also provide a computer readable storage medium. The above-mentioned method according to the embodiments of the present application can be implemented in hardware, firmware, or as computer code recorded on a storage medium, or be implemented by downloading a computer program from a network and stored in a remote storage medium or a non-transitory machine readable storage medium and stored in a local storage medium, so that the method described herein can be stored on a storage medium by such software processing using a general purpose computer, a special purpose processor, or programmable or special purpose hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, and the like. Further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that the computer, the processor, the microprocessor controller, or the programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above-mentioned embodiments is implemented.

[0207] Part of the present application can be applied as a computer program product, for example, computer program instructions, when executed by a computer, through the operation of the computer, can invoke or provide the method and / or technical solutions according to the present application. Those skilled in the art should understand that the form of computer program instructions in computer readable medium includes but is not limited to source files, executable files, installation package files and the like, and accordingly, the way of computer program instructions executed by computer includes but is not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Here, the computer readable medium can be any available computer readable storage medium or communication medium accessible to the computer.

[0208] Although the embodiments of the present application are described in conjunction with the drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. An active seat belt control method characterized by, The method comprises: receiving different monitoring information sent by multiple control domains, obtaining global awareness information based on the monitoring information; extracting local information from the global awareness information according to different preset scenes to obtain target awareness information corresponding to each preset scene; performing fuzzy reasoning on the target awareness information to obtain a risk level corresponding to each preset scene, and generating an active safety belt control instruction corresponding to each preset scene; determining a target active safety belt control instruction based on the risk level and the active safety belt control instruction, and executing the target active safety belt control instruction; the fuzzy reasoning on the target awareness information to obtain a risk level corresponding to each preset scene comprises: performing fuzzy processing on the target awareness information to obtain fuzzy awareness information; determining a fuzzy output set corresponding to each preset scene for representing a risk level according to a preset fuzzy control rule corresponding to each preset scene and the fuzzy awareness information; the fuzzy control rule is constructed according to a group of preset fuzzy condition sentences; the fuzzy output corresponding to the fuzzy awareness information is obtained by using the fuzzy control rule to comprehensively evaluate a series of logical conditions of the fuzzy awareness information; the fuzzy reasoning calculation process of each preset scene is based on all the fuzzy awareness information corresponding to the preset scene; when a certain fuzzy control rule is activated, the corresponding fuzzy output is obtained, and finally the fuzzy output set including multiple fuzzy outputs is obtained; obtaining a risk level corresponding to each preset scene based on the fuzzy output set; the obtaining of a risk level corresponding to each preset scene based on the fuzzy output set comprises: for each preset scene, performing defuzzification processing on multiple fuzzy outputs in the fuzzy output set corresponding to the current preset scene to obtain a quantitative output corresponding to each fuzzy output; for each preset scene, performing weighted summation on multiple quantitative outputs corresponding to the current preset scene to obtain a risk level corresponding to the current preset scene.

2. The method of claim 1, wherein, the determination of a target active safety belt control instruction based on the risk level and the active safety belt control instruction comprises: determining a target preset scene with the maximum risk level based on the risk level corresponding to each preset scene; determining a target active safety belt control instruction according to the active safety belt control instruction corresponding to the target preset scene.

3. The method according to claim 1 or 2, characterized in that, before executing the target active safety belt control instruction, the method further comprises: obtaining state information of a safety belt hardware execution mechanism; wherein the state information comprises at least one of network communication state, motor power-on state, motor start-up and running state, safety belt pre-execution action signal level, safety belt pre-execution action time length, safety belt execution action abnormal information, remaining available number of safety belt execution action signals, and degradation execution information of safety belt execution action; performing state evaluation on the safety belt hardware execution mechanism based on the state information, and when the state evaluation passes, jumping to the step of executing the target active safety belt control instruction.

4. The method of claim 1, wherein, the obtaining of global awareness information based on the monitoring information comprises: The monitoring information is subjected to validity verification, resampling and numerical transformation to obtain global perception information; wherein the monitoring information comprises at least two of vehicle motion information, chassis control information, advanced driving assistance control information, road environment information, driver and passenger state information, active driving comfort control information and collision prediction information.

5. The method of claim 4, wherein, The preset scenarios comprise at least two of a comfort improvement scenario, a safety reminder scenario, a dynamic support scenario, an emergency protection scenario and a collision protection scenario; the local information extraction of the global perception information according to different preset scenarios to obtain target perception information corresponding to each preset scenario comprises: For the comfort improvement scenario, the driver and passenger state information or the active driving comfort control information in the global perception information is extracted to obtain target perception information corresponding to the comfort improvement scenario; and / or, for the safety reminder scenario, the driver and passenger state information or the advanced driving assistance control information in the global perception information is extracted to obtain target perception information corresponding to the safety reminder scenario; and / or, for the dynamic support scenario, the vehicle motion information, the chassis control information, the road environment information, the driver and passenger state information or the collision prediction information in the global perception information is extracted to obtain target perception information corresponding to the dynamic support scenario; and / or, for the emergency protection scenario, the vehicle motion information, the chassis control information, the advanced driving assistance control information, the road environment information or the driver and passenger state information in the global perception information is extracted to obtain target perception information corresponding to the emergency protection scenario; and / or, for the collision protection scenario, the driver and passenger state information or the collision prediction information in the global perception information is extracted to obtain target perception information corresponding to the collision protection scenario.

6. The method of claim 5, wherein, The fuzzy inference of the target perception information generates active seatbelt control instructions corresponding to each preset scenario, comprising: fuzzy inference based on the target perception information corresponding to the comfort improvement scenario to generate a roll-back instruction; and / or, fuzzy inference based on the target perception information corresponding to the safety reminder scenario to generate a reminder instruction; and / or, fuzzy inference based on the target perception information corresponding to the dynamic support scenario to generate a tightening instruction; and / or, fuzzy inference based on the target perception information corresponding to the emergency protection scenario to generate a pull-back instruction; and / or, fuzzy inference based on the target perception information corresponding to the collision protection scenario to generate a warning instruction.

7. An active seat belt control device characterized by comprising: The device comprises: an acquisition module for receiving different monitoring information sent by multiple control domains and obtaining global perception information based on the monitoring information; a first processing module for local information extraction of the global perception information according to different preset scenarios to obtain target perception information corresponding to each preset scenario; a second processing module for fuzzy inference of the target perception information to obtain a risk level corresponding to each preset scenario and generate active seatbelt control instructions corresponding to each preset scenario; a control module for determining a target active seatbelt control instruction based on the risk level and the active seatbelt control instructions and executing the target active seatbelt control instruction; the second processing module is further configured to: The target perception information is fuzzified to obtain fuzzy perception information; According to the preset fuzzy control rules and the fuzzy perception information corresponding to different preset scenes, a fuzzy output set corresponding to the risk level of each preset scene is determined; a set of fuzzy condition statements is preset to constitute the fuzzy control rules, and the fuzzy perception information is comprehensively evaluated through a series of logical conditions using the fuzzy control rules to obtain the fuzzy output corresponding to the fuzzy perception information; the fuzzy inference calculation process of each preset scene is based on all the fuzzy perception information corresponding thereto, when a certain rule in the fuzzy control rules is activated, the corresponding fuzzy output is obtained, and finally the fuzzy output set including multiple fuzzy outputs is obtained; Based on the fuzzy output set, the risk level corresponding to each preset scene is obtained; The second processing module is further configured to: For each preset scene, the multiple fuzzy outputs in the fuzzy output set corresponding to the current preset scene are respectively defuzzified to obtain quantized outputs corresponding to the fuzzy outputs; For each preset scene, the multiple quantized outputs corresponding to the current preset scene are weighted and summed to obtain the risk level corresponding to the current preset scene.

8. A vehicle characterized by comprising: It comprises: a memory and a processor, which are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the active seat belt control method of any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for causing a computer to perform the active seat belt control method of any one of claims 1 to 6.

10. A computer program product, characterised in that, It comprises computer instructions for causing a computer to perform the active seat belt control method of any one of claims 1 to 6.

Images