Vehicle floating water stage identification method, device and equipment, vehicle and storage medium

Abstract

Embodiments of the present application provide a vehicle floating water stage identification method, device, equipment, vehicle and storage medium. The method comprises: based on the data collected by each sensor of the vehicle, constructing at least two information sources, the at least two information sources comprising at least two of the following: a vehicle dynamics information source, a kinematics information source, a water depth information source, a suspension information source and a front image information source; performing feature extraction on the at least two information sources respectively to generate an independent confidence corresponding to each floating water stage for each information source, the floating water stage comprising an entry stage, a floating stage and an exit stage; fusing each independent confidence to generate a comprehensive confidence; and identifying the floating water stage in which the vehicle is currently located based on the comprehensive confidence. The method is used to accurately identify the floating water stage in which the vehicle is located when the vehicle is wading.

Description

Technical Field

[0001] This application relates to the field of intelligent vehicle technology, and in particular to a method, apparatus, device, vehicle, and storage medium for identifying the floating stage of a vehicle. Background Technology

[0002] Vehicle wading identification typically relies on sensor information such as water depth, wheel speed, suspension, or vision, and uses threshold judgment or rule fusion to identify whether a vehicle is in a floating state.

[0003] However, existing solutions are mostly based on raw sensor signals and fixed rules for judgment, which are easily affected by murky water, low light, slippage and water level fluctuations. When the sensor is abnormal or there is information conflict, it is difficult to accurately represent the state change, resulting in poor recognition accuracy.

[0004] Therefore, improving the accuracy of vehicle buoyancy recognition in complex wading environments has become an urgent technical problem to be solved. Summary of the Invention

[0005] The vehicle floating stage identification method, apparatus, equipment, vehicle, and storage medium provided in this application are used to improve the accuracy of vehicle floating stage identification.

[0006] In a first aspect, embodiments of this application provide a method for identifying the floating stage of a vehicle, including:

[0007] Based on the data collected by various sensors of the vehicle, at least two information sources are constructed, including at least two of the following: vehicle dynamics information source, kinematics information source, water depth information source, suspension information source, and forward image information source.

[0008] For each floating stage, feature extraction is performed on the at least two information sources to generate an independent confidence level for each information source in each floating stage. The independent confidence level of any information source in the floating stage is used to characterize the probability that the vehicle is in the floating stage as determined by the information source.

[0009] The independent confidence scores of each information source at the same floating stage are merged to generate a comprehensive confidence score for each floating stage.

[0010] Based on the overall confidence level of each floating stage, the floating stage of the vehicle at the current moment is identified, and the floating stage includes the entry stage, the floating stage, or the exit stage.

[0011] In one possible implementation, identifying the current floating stage of the vehicle based on the overall confidence level of each floating stage includes:

[0012] The target comprehensive confidence score is determined from the various comprehensive confidence scores, wherein the target comprehensive confidence score is the largest value among the various comprehensive confidence scores;

[0013] The floating stage corresponding to the target comprehensive confidence level is determined as the floating stage of the vehicle at the current moment.

[0014] In one possible implementation, the step of fusing the independent confidence levels of various information sources corresponding to the same floating stage to generate a comprehensive confidence level for each floating stage includes:

[0015] Obtain the uncertainty ratio coefficient corresponding to each information source;

[0016] Based on the uncertainty ratio coefficient corresponding to each information source, the independent confidence level of each information source in each floating stage, and the sum of independent confidence levels of each information source in each floating stage, the basic probability distribution of each information source for each floating stage is calculated.

[0017] Based on the DS evidence theory, the basic probability distribution of each information source for each floating stage is probabilistically fused to generate a comprehensive confidence level for each floating stage.

[0018] In one possible implementation, the step of extracting features from the at least two information sources and generating an independent confidence level for each information source at each floating stage includes:

[0019] If the dynamic information source is included among the at least two information sources, then the vehicle driving force, vehicle motion resistance and actual acceleration are extracted from the dynamic information source.

[0020] Calculate the actual acceleration drag based on the vehicle mass and the actual acceleration;

[0021] Based on the vehicle driving force, the vehicle motion resistance, and the vehicle dynamics equations, calculate the theoretical expected acceleration resistance;

[0022] Based on the actual acceleration drag and the theoretical expected acceleration drag, calculate the acceleration drag difference and the rate of change of the acceleration drag difference;

[0023] Based on the acceleration resistance difference and the rate of change of the acceleration resistance difference, the independent confidence level of the kinetic information source is determined for each floating stage.

[0024] In one possible implementation, determining the independent confidence level of the kinetic information source for each floating stage based on the acceleration drag difference and the rate of change of the acceleration drag difference includes:

[0025] When the rate of change of the acceleration resistance difference is greater than zero, the minimum value is selected from the ratio of the rate of change of the acceleration resistance difference to the first slope and the preset confidence upper limit value, and is used as the independent confidence level of the dynamic information source in the water entry stage; the first slope is the slope threshold of the rate of change of the acceleration resistance difference when it increases;

[0026] When the rate of change of the acceleration resistance difference is less than or equal to zero, the preset confidence lower limit value will be used as the independent confidence level of the dynamic information source in the water entry stage.

[0027] The minimum value is selected from the ratio of the acceleration resistance difference to the preset force deviation threshold and the preset confidence upper limit value, and is used as the independent confidence level of the dynamic information source in the floating stage.

[0028] When the rate of change of the acceleration resistance difference is greater than zero, the minimum value is selected from the ratio of the rate of change of the acceleration resistance difference to the second slope and the preset confidence upper limit value, and is used as the independent confidence level of the dynamic information source in the water effluent stage; the second slope is the slope threshold of the rate of change of the acceleration resistance difference when it decreases.

[0029] When the rate of change of the acceleration resistance difference is less than or equal to zero, the preset confidence lower limit is used as the independent confidence level of the dynamic information source in the water discharge stage.

[0030] In one possible implementation, the step of extracting features from the at least two information sources and generating an independent confidence level for each information source at each floating stage includes:

[0031] If the kinematic information source is included among the at least two information sources, then the wheel speed difference between the front axle and the rear axle of the vehicle, the front axle slip ratio, and the rear axle slip ratio of the vehicle are extracted from the kinematic information source.

[0032] Based on the wheel speed difference, the vehicle's front axle slip ratio, and the vehicle's rear axle slip ratio, the independent confidence level of the kinematic information source is determined for each floating stage.

[0033] In one possible implementation, determining the independent confidence level of the kinematic information source for each floating stage based on the wheel speed difference, the vehicle's front axle slip ratio, and the vehicle's rear axle slip ratio includes:

[0034] Based on the rate of change of wheel speed difference, the lower limit threshold of the rate of change of wheel speed difference, the rate of change of the vehicle front axle slip ratio, the lower limit threshold of the rate of change of the front axle slip ratio, the preset first weighting coefficient, and the preset upper limit of confidence, the independent confidence level of the kinematic information source during the water entry phase is calculated.

[0035] Based on the wheel speed difference, the wheel speed difference change rate threshold, the vehicle front axle slip rate, the vehicle rear axle slip rate, the front axle slip rate change rate threshold, the preset second weighting coefficient, and the preset confidence upper limit value, calculate the independent confidence level of the kinematic information source during the floating phase;

[0036] Based on the rate of change of wheel speed difference, the upper limit threshold of the rate of change of wheel speed difference, the rate of change of the vehicle's front axle slip ratio, the upper limit threshold of the rate of change of the front axle slip ratio, the preset third weighting coefficient, and the preset confidence upper limit value, the independent confidence level of the kinematic information source during the water exit phase is calculated.

[0037] In one possible implementation, the step of extracting features from the at least two information sources and generating an independent confidence level for each information source at each floating stage includes:

[0038] If the water depth information source is included among the at least two information sources, then the wading depth on the left side of the vehicle and the wading depth on the right side of the vehicle are extracted from the water depth information source.

[0039] Based on the wading depths on the left and right sides of the vehicle, calculate the mean water depth and the rate of change of water depth.

[0040] Based on the mean water depth and the rate of change of water depth, the independent confidence level of the water depth information source is determined for each floating stage.

[0041] In one possible implementation, determining the independent confidence level of the water depth information source for each floating stage based on the mean water depth and the rate of change of water depth includes:

[0042] Based on the water depth change rate, the water depth mean, the weighting coefficient of the water depth change rate during the water entry stage, the water depth weighting coefficient during the water entry stage, the water depth threshold for vehicle water entry, the water depth rise rate threshold, and the preset confidence upper limit, the independent confidence level of the water depth information source during the water entry stage is calculated.

[0043] Based on the absolute value of the water depth change rate, the mean water depth, the weighting coefficient of the water depth change rate during the floating phase, the weighting coefficient of the water depth during the floating phase, the allowable rate of water depth fluctuation during the floating phase, the installation height of the vehicle's water depth sensor above the ground, and the preset confidence limit value, the independent confidence level of the water depth information source during the floating phase is calculated.

[0044] Based on the water depth change rate, the mean water depth, the weighting coefficient of the water depth change rate during the water discharge stage, the water depth weighting coefficient during the water discharge stage, the water depth threshold for vehicle water discharge, the water depth descent rate threshold, and the preset confidence upper limit, the independent confidence level of the water depth information source during the water discharge stage is calculated.

[0045] In one possible implementation, the step of extracting features from the at least two information sources and generating an independent confidence level for each information source at each floating stage includes:

[0046] If the suspension information source is included among the at least two information sources, then the vehicle front axle suspension holding force, the rate of change of the front axle suspension holding force, the vehicle rear axle suspension holding force, and the rate of change of the rear axle suspension holding force are extracted from the suspension information source.

[0047] Based on the vehicle's front axle suspension holding force, the rate of change of the front axle suspension holding force, the vehicle's rear axle suspension holding force, and the rate of change of the rear axle suspension holding force, the independent confidence level of the suspension information source corresponding to each floating stage is determined.

[0048] In one possible implementation, determining the independent confidence level of the suspension information source for each floating stage based on the vehicle's front axle suspension holding force, the rate of change of the front axle suspension holding force, the vehicle's rear axle suspension holding force, and the rate of change of the rear axle suspension holding force includes:

[0049] Based on the vehicle's front axle suspension holding force, the rate of change of the front axle suspension holding force, the threshold of the rate of decrease of the front axle suspension holding force, the weighting coefficient of the rate of change of the front axle suspension holding force during the water entry stage, the weighting coefficient of the front axle suspension holding force, the front axle suspension holding force when the vehicle is stationary on a level road surface, and the preset confidence upper limit, the independent confidence level of the suspension information source during the water entry stage is calculated.

[0050] Based on the vehicle's front axle suspension holding force, the vehicle's rear axle suspension holding force, the vehicle's front axle suspension holding force when stationary on a level road surface, the vehicle's rear axle suspension holding force when stationary on a level road surface, the weighting coefficient of the front axle suspension holding force during the floating phase, the weighting coefficient of the rear axle suspension holding force during the floating phase, and the preset confidence upper limit value, calculate the independent confidence level of the suspension information source during the floating phase.

[0051] Based on the front axle suspension holding force change rate, the vehicle's front axle suspension holding force, the vehicle's front axle suspension holding force when stationary on a level road surface, the front axle suspension holding force increase rate threshold, the weighting coefficient of the front axle suspension holding force change rate during the water exit phase, the weighting coefficient of the front axle suspension holding force during the water exit phase, and the preset confidence upper limit value, the independent confidence level of the suspension information source during the water exit phase is calculated.

[0052] In one possible implementation, the step of extracting features from the at least two information sources and generating an independent confidence level for each information source at each floating stage includes:

[0053] If the forward image information source is included among the at least two information sources, then the forward image of the vehicle is extracted from the forward image information source;

[0054] Acquire the water surface area and road surface area in the image in front of the vehicle;

[0055] Based on the water surface area and the road surface area, calculate the proportion of the water-involved area and the rate of change of the proportion of the water-involved area;

[0056] Based on the proportion of the water-touched area and the rate of change of the proportion of the water-touched area, the independent confidence level of the forward image information source is determined for each floating stage.

[0057] In one possible implementation, determining the independent confidence level of the foreground image information source at each floating stage based on the proportion of the wading area and the rate of change of the proportion of the wading area includes:

[0058] Based on the rate of change of the proportion of the wading area, the proportion of the wading area, the weight of the rate of change of the proportion of the wading area in the front image of the vehicle during the water entry stage, the threshold for the increase of the rate of change of the proportion of the wading area, the weight of the proportion of the wading area, the saturation threshold of the proportion of the wading area and the preset confidence upper limit, the independent confidence of the front image information source during the water entry stage is calculated.

[0059] Based on the proportion of the wading area, the weight of the proportion of the wading area in the vehicle's front image during the floating phase, the high proportion threshold of the wading area, and the preset confidence upper limit, the independent confidence level of the front image information source during the floating phase is calculated.

[0060] Based on the change rate of the proportion of the wading area, the proportion of the wading area, the weight of the change rate of the proportion of the wading area in the vehicle's front image during the water exit stage, the threshold for the decrease in the change rate of the proportion of the wading area, the weight of the proportion of the wading area, the threshold for the low proportion of the wading area, and the preset confidence upper limit, the independent confidence level of the front image information source during the water exit stage is calculated.

[0061] In one possible implementation, the method further includes:

[0062] The sensor signals collected by each sensor of the vehicle are acquired, including water depth signal, vehicle suspension holding force signal, vehicle front and rear axle wheel speed difference signal, front and rear axle wheel speed change signal, and vehicle pitch angle signal.

[0063] Based on at least one of the water depth signal, the vehicle suspension holding force signal, the vehicle front and rear axle wheel speed difference signal, and the front and rear axle wheel speed change signal, the degree of water penetration of the vehicle's front axle is determined, and the degree of water penetration is divided into front axle not submerged, front axle slightly submerged, and front axle deeply submerged.

[0064] The floating attitude of the vehicle during the floating phase is identified based at least on the degree of water entry of the front axle of the vehicle and the vehicle pitch angle signal.

[0065] Secondly, embodiments of this application provide a vehicle floating stage identification device, comprising:

[0066] A construction module is used to construct at least two information sources based on data collected by various sensors of the vehicle. The at least two information sources include at least two of the following: vehicle dynamics information source, kinematics information source, water depth information source, suspension information source, and forward image information source.

[0067] An extraction module is used to extract features from the at least two information sources for each floating stage, and generate an independent confidence score for each information source for each floating stage. The independent confidence score of any information source for the floating stage is used to characterize the probability that the vehicle is in the floating stage as determined by the information source.

[0068] The generation module is used to merge the independent confidence scores of various information sources at the same floating stage to generate a comprehensive confidence score for each floating stage.

[0069] The identification module is used to identify the current floating stage of the vehicle based on the comprehensive confidence level of each floating stage, wherein the floating stage includes the water entry stage, the floating stage, or the water exit stage.

[0070] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0071] The memory stores computer-executed instructions;

[0072] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0073] Fourthly, embodiments of this application provide a vehicle, including a vehicle body and the aforementioned electronic equipment.

[0074] Fifthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0075] The vehicle floating stage identification method, device, equipment, vehicle, and storage medium provided in this application embodiment calculate the independent confidence levels of at least two information sources, such as dynamic information sources, kinematic information sources, water depth information sources, suspension information sources, and forward image information sources, at each floating stage, and then fuse them to generate a comprehensive confidence level, ultimately outputting the current floating stage of the vehicle. This can effectively reduce the impact of single sensor anomalies on the identification results in wading scenarios with high interference and high uncertainty, and improve the accuracy of vehicle floating stage identification. Attached Figure Description

[0076] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0077] Figure 1 This is a schematic diagram of the vehicle floating stage identification method provided in the embodiments of this application;

[0078] Figure 2 This is a schematic diagram of the vehicle floating stage provided in an embodiment of this application;

[0079] Figure 3 Flowchart of the method for generating independent confidence scores for each floating stage of the kinetic information source provided in this application embodiment;

[0080] Figure 4 This is a schematic diagram of the longitudinal forces acting on a vehicle, provided in an embodiment of this application.

[0081] Figure 5 A flowchart illustrating the generation of independent confidence levels for each floating stage of the water depth information source provided in this application embodiment;

[0082] Figure 6 A flowchart illustrating the generation of independent confidence levels for each floating stage of the suspension information source provided in this application embodiment;

[0083] Figure 7 A flowchart illustrating the generation of independent confidence levels for each floating stage of the foreground image information source provided in this application embodiment;

[0084] Figure 8 This is a schematic diagram illustrating the changes in images acquired during the vehicle's entry into water phase, provided in an embodiment of this application.

[0085] Figure 9 This is a schematic diagram illustrating the changes in images acquired during the vehicle's water exit phase, as provided in an embodiment of this application.

[0086] Figure 10 This is a schematic diagram of the vehicle floating stage identification framework provided in an embodiment of this application;

[0087] Figure 11This is a schematic diagram of the vehicle's floating posture provided in an embodiment of this application;

[0088] Figure 12 A schematic diagram of the vehicle floating stage identification device provided in this application;

[0089] Figure 13 A schematic diagram of the structure of the electronic device provided in this application.

[0090] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0091] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0092] As vehicle intelligence levels continue to improve and driving scenarios expand, users can utilize water-driving capabilities to cope with urban flooding, wading through water, and other similar situations. However, when the wading depth exceeds a safe threshold, users face an extremely high risk of the vehicle transitioning from road driving to floating in the water. In this floating state, the adhesion between the tires and the ground is completely lost, the vehicle's traditional dynamics model becomes ineffective, and it is highly likely to cause the vehicle to lose control, roll over, or sink, posing a serious threat to the driver and passengers. Therefore, developing a fusion system capable of accurately identifying the vehicle's floating stage in real time, providing precise decision-making basis for intelligent control of vehicle water driving, has become a core prerequisite for improving vehicle water-wading safety.

[0093] In related technologies, the following strategies are mainly used to achieve floating detection: (1) Floating detection scheme based on single sensor threshold: It relies on a single type of sensor signal, such as the vehicle wading depth signal collected by the water depth sensor, and judges whether the vehicle has entered the floating state by setting a fixed threshold. It is easily affected by specific interference and will produce misjudgment; (2) Simple logic judgment scheme based on multiple sensors: It uses a simple logic or static weighted average method to combine and judge multiple sensor signals. When the sensor signals are abnormal or contradictory, the system can only rely on the default strategy and cannot achieve adaptive adjustment. The reliability and robustness are low.

[0094] To address the aforementioned issues, this application provides a method, apparatus, device, vehicle, and storage medium for identifying the floating stage of a vehicle. The method constructs at least two information sources based on sensor data collected by various sensors of the vehicle, enabling information from different sources to jointly characterize the vehicle's current wading state from multiple perspectives. Subsequently, features are extracted from each information source to generate an independent confidence level for each information source at each floating stage, thereby transforming the originally scattered, heterogeneous, and easily interfered sensor data into a unified expression usable for stage determination. Then, the independent confidence levels are fused to generate a comprehensive confidence level, reducing the adverse impact of anomalies or conflicts from a single information source on the identification results. Finally, the floating stage of the vehicle at the current moment is identified based on the comprehensive confidence level, thus providing a more reliable floating stage identification capability for vehicles in wading scenarios with high interference and high uncertainty.

[0095] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0096] Figure 1 This is a schematic flowchart of a vehicle floating stage identification method provided in an embodiment of this application. This method can be deployed in a perception and analysis architecture composed of multiple types of onboard sensors and onboard processing units. By organizing, refining, and fusing multiple information sources, it achieves dynamic identification of the vehicle's floating stage. Figure 1 As shown, the method includes:

[0097] Step 110: Based on the data collected by the various sensors of the vehicle, construct at least two information sources; wherein, the at least two information sources include at least two of the following: vehicle dynamics information source, kinematics information source, water depth information source, suspension information source, and forward image information source;

[0098] Step 120: For each floating stage, feature extraction is performed on at least two information sources to generate an independent confidence level for each information source in each floating stage; wherein, the independent confidence level of any information source in the floating stage is used to characterize the probability that the vehicle is in the floating stage as determined by the information source.

[0099] Step 130: Merge the independent confidence scores of each information source at the same floating stage to generate a comprehensive confidence score for each floating stage;

[0100] Step 140: Based on the overall confidence level of each floating stage, identify the floating stage in which the vehicle is currently at the current moment; wherein, the floating stage includes the water entry stage, the floating stage, or the water exit stage.

[0101] For step 110 above, the sensors on the vehicle may include at least: (1) an inertial measurement unit (IMU) for measuring the longitudinal acceleration of the vehicle to calculate the actual acceleration resistance of the vehicle; (2) a camera for acquiring images of the front of the vehicle; (3) a water depth sensor and a pressure sensor for acquiring the wading depth of the vehicle; (4) a wheel speed sensor for acquiring the wheel speed; (5) an on-board positioning sensor for acquiring the actual vehicle speed; and (6) a force sensor for acquiring the front axle suspension holding force and the rear axle suspension holding force of the vehicle.

[0102] In complex floating environments, any single type of sensor has inherent limitations. For example, cameras experience a sharp decline in performance in murky water, when obscured by splashes, or at night; wheel speed sensors fail to register signals when the tires are completely off the ground and slipping; and traditional floating detection solutions rely on a particular type of sensor. If the working environment of that sensor deteriorates, the vehicle's entire perception system is prone to misjudging the current floating stage, leading to completely incorrect vehicle control strategies and causing serious safety hazards.

[0103] In this embodiment, the sensor data collected by each sensor can be divided into different information sources according to the state dimension.

[0104] For example, sensor data can be divided into dynamic information sources, kinematic information sources, water depth information sources, suspension information sources, and forward image information sources.

[0105] The dynamic information source is used to characterize the state information of a vehicle under the influence of driving force, motion resistance and acceleration changes during water wading, and reflects the dynamic response characteristics of the vehicle in the water entry, floating and exit stages.

[0106] Kinematic information sources are used to characterize vehicle wheel speed and slip state information. They can be generated from wheel speed sensors, wheel angle information, and vehicle speed estimation results to depict the difference in wheel speed between the front and rear axles and slip changes.

[0107] Water depth information sources are used to characterize the wading depth information around a vehicle. They can be obtained by water level sensors deployed on the side, front or bottom of the vehicle body, reflecting the local water depth changes on the left and right sides of the vehicle.

[0108] Suspension information sources are used to characterize the changes in support forces between the vehicle body and the wheels. They can be constructed from suspension travel sensors, pressure sensors, or height sensors to reflect the holding forces and their rates of change of the front and rear axle suspensions of the vehicle.

[0109] A forward image information source is used to characterize the appearance of the water surface and road surface in front of the vehicle. It can be collected by a camera installed at the front of the vehicle, and the proportion of the wading area and its changing characteristics can be formed through image processing.

[0110] In practical applications, the aforementioned information sources can be connected to in-vehicle processing equipment with data processing capabilities, and the in-vehicle processing equipment will perform subsequent processing in a unified manner.

[0111] Among them, the dynamic information source can be understood as the set of information used to determine the floating stage from the perspective of vehicle dynamics; the kinematic information source can be understood as the set of information used to determine the floating stage from the perspective of wheel speed and slippage; the water depth information source can be understood as the set of information used to determine the floating stage from the perspective of wading depth; the suspension information source can be understood as the set of information used to determine the floating stage from the perspective of changes in suspension holding force; and the forward image information source can be understood as the set of information used to determine the floating stage from the perspective of changes in the visual features of the front of the vehicle.

[0112] For step 120 above, feature extraction is used to extract key quantities that are directly related to the identification of the floating stage from various information sources, and to further map these key quantities to the degree of support (i.e., independent confidence) in the water entry stage, floating stage and water exit stage.

[0113] Independent confidence is used to represent the strength of support of a single information source for a certain floating stage at a certain moment. It is a comparable and fusionable stage judgment quantity, which can be represented by a value between 0 and 1.

[0114] The higher the independent confidence score, the more the corresponding information source supports the floating stage. For example, if the independent confidence score of the dynamic information source is 0.8 for the water entry stage, 0.6 for the floating stage, and 0.2 for the water exit stage, then the dynamic information source strongly supports that the vehicle is currently in the water entry stage.

[0115] The water entry phase corresponds to the process where the vehicle gradually comes into contact with and enters the water from normal driving, buoyancy begins to increase, and adhesion has not yet been completely lost. The floating phase corresponds to the process where the vehicle is significantly affected by buoyancy, the tire adhesion ability is significantly reduced, and the vehicle may even lose effective support. The water exit phase corresponds to the process where the vehicle gradually leaves the high water level area, adhesion is restored, and the effect of buoyancy weakens.

[0116] Traditional floating detection methods cannot extract high-order feature signals strongly correlated with the three floating stages from low-order sensor data. The raw sensor data (such as water depth signals) are low-order signals, making it difficult to extract decisive features that clearly define a vehicle's entry into water, floating, and exit from the water. For example, a simple increase in water depth could be due to a vehicle passing through a flooded section of road or rapidly plunging into deep water. Traditional floating detection methods lack in-depth processing and feature extraction of this raw data, thus failing to achieve high-precision stage identification.

[0117] In this embodiment, an advanced feature extraction algorithm is tailored for each sensor source, and its independent confidence vectors for the three floating stages are output. This enables continuous and accurate identification of the three dynamic stages of entering the water, floating, and exiting the water, providing a more refined decision-making basis for vehicle control.

[0118] Regarding step 130 above, the overall confidence level is the degree of joint support from multiple information sources for the water entry stage, floating stage, and water exit stage, and serves as a unified decision-making basis for finally identifying the vehicle's floating stage.

[0119] In practical applications, different information sources are affected by environmental disturbances in different ways. At the same time, there may be inconsistencies such as the front image information source being more supportive of the water entry phase and the suspension information source being more supportive of the floating phase.

[0120] Traditional multi-source information fusion schemes mainly use weighted average algorithms, which cannot handle conflicting sensor signals under extreme conditions, causing misidentification during the floating stage. This leads to dangerous decisions by the downstream vehicle control system. Therefore, the robustness and reliability of traditional fusion methods do not meet the requirements of vehicle floating function.

[0121] This embodiment employs a confidence fusion mechanism to coordinate and process each independent confidence level, thereby reducing the adverse impact of abnormal information sources, distorted information sources, or low-reliability information sources on the final judgment.

[0122] For example, the Dempster-Shafer (DS) evidence theory can be used as the core of the fusion algorithm to achieve the fusion of multiple independent confidence levels.

[0123] In other implementations, firstly, the uncertainty ratio coefficient corresponding to each information source can be obtained; then, based on the uncertainty ratio coefficient corresponding to each information source, the independent confidence level of each information source at each floating stage, and the sum of the independent confidence levels of each information source at each floating stage, the basic probability distribution of each information source at each floating stage is calculated; finally, based on the DS evidence theory, the basic probability distribution of each information source at each floating stage is probabilistically fused to generate the comprehensive confidence level for each floating stage.

[0124] In this embodiment, before performing probability fusion on the basic probability distribution of each floating stage, the on-board processing device can first generate an uncertainty ratio coefficient for each information source. This uncertainty ratio coefficient is used to represent the reliability of the information source in its current state. Specifically, the uncertainty ratio coefficient can be calculated based on the sensor health status, data fluctuation, feature self-consistency, and historical stability. The larger the value of the uncertainty ratio coefficient, the higher the current uncertainty of the information source, and the less weight it is allocated in the subsequent fusion.

[0125] In this embodiment, the DS evidence theory identification framework is the set of all possible outcomes. The identification framework can be defined as follows: .in, Indicates the water entry phase. Indicates the floating phase. This indicates the water discharge stage.

[0126] The DS evidence theory requires fusing the basic probability distributions provided by multiple information sources to generate a comprehensive probability allocation. This strengthens the confidence of propositions supported by multiple information sources and weakens the confidence of propositions supported by a few information sources. Each of the five information sources outputs a three-dimensional confidence vector:

[0127]

[0128] In the above formula, =1 represents the source of dynamic information. =2 represents the kinematic information source. =3 represents the water depth information source. =4 represents the suspension information source. =5 represents the source of the image information in front. Indicates the first A three-dimensional confidence vector is composed of the independent confidence levels of each information source during the water entry, floating, and water exit phases.

[0129] Because the DS evidence theory requires that the basic probability distribution satisfies Therefore, probability normalization is required for each confidence vector. The basic probability calculation for each information source is as follows:

[0130]

[0131]

[0132]

[0133] In the above formula, This represents the basic probability of the i-th information source during the water entry phase. This represents the basic probability of the i-th information source during the floating phase. Let represent the basic probability of the i-th information source during the water discharge stage. , This represents the sum of independent confidence levels for the i-th information source at each stage of the floating phase. For the first The uncertainty ratio coefficient corresponding to each information source can be dynamically adjusted according to the reliability of the sensor. For example, if the reliability of a camera decreases at night, its corresponding uncertainty ratio coefficient will increase.

[0134] After normalization, probability fusion of multiple information sources is performed based on the following rules:

[0135]

[0136] In the above formula, All are recognition frameworks a subset of The conflict coefficient is used to measure the degree of conflict between information sources.

[0137]

[0138] The basic probabilities of the five information sources mentioned above are fused pairwise, in the following order:

[0139]

[0140] In summary, the basic probability after fusion based on DS evidence theory is obtained. That is, the overall confidence level.

[0141] Regarding step 140 above, the stage identification based on comprehensive confidence is used to output which state the vehicle is currently in: the water entry stage, the floating stage, or the water exit stage.

[0142] In some implementations, a target comprehensive confidence level can be determined from various comprehensive confidence levels; the floating stage corresponding to the target comprehensive confidence level is determined as the floating stage the vehicle is in at the current moment. The target comprehensive confidence level is the highest value among the various comprehensive confidence levels.

[0143] For example, if the overall confidence level of the floating stage is greater than the overall confidence level of the water entry stage and the overall confidence level of the water exit stage for N consecutive periods, and its value exceeds the floating identification threshold, then the vehicle is identified as being in the floating stage; if the overall confidence level of the water entry stage continues to rise in a short period of time and exceeds the water entry identification threshold, while the overall confidence level of the floating stage has not yet formed a stable advantage, then the vehicle is identified as being in the water entry stage; if the overall confidence level of the water exit stage is consistently higher than that of other stages and is accompanied by consistent trend signals such as water depth decrease and suspension holding force recovery, then the vehicle is identified as being in the water exit stage.

[0144] Additionally, in some other implementations, the fused base probabilities can be compared. For three singleton sets The support (i.e., the combined confidence of the three floating stages) is used to select the set of elements with the highest support as the final output of the vehicle floating stage judgment.

[0145] In this embodiment, the final decision can be made based on the following criteria:

[0146]

[0147] In the above formula, For inclusion The set, For the maximum confidence criterion function, The final output is the floating stage, which is the vehicle floating stage output after multi-source information fusion based on DS evidence theory.

[0148] The vehicle floating stage identification method provided in this application utilizes the independent confidence levels of at least two information sources, such as dynamic information sources, kinematic information sources, water depth information sources, suspension information sources, and forward image information sources, at each floating stage, and then fuses them to generate a comprehensive confidence level, ultimately outputting the current floating stage of the vehicle. This method can effectively reduce the impact of single sensor anomalies on the identification results in wading scenarios with high interference and high uncertainty, enhance the continuous representation capability of the stage evolution process, and thus improve the accuracy, stability, and robustness of vehicle floating state identification.

[0149] The following examples detail how to generate independent confidence levels for each information source at each floating stage.

[0150] First, this application divides the vehicle's floating phase into three processes: the water entry phase, the floating phase, and the water exit phase. Figure 2 This is a schematic diagram of the vehicle floating stage provided in an embodiment of this application, such as... Figure 2 As shown, this embodiment of the application will identify the three floating stages during the floating process from five aspects: vehicle dynamics, vehicle kinematics, water depth signals detected by water depth sensors, suspension height holding force signals, and image signals acquired by cameras. It will then output the independent confidence scores corresponding to the vehicle in each of the three floating stages, namely, the confidence score for the water entry stage. Confidence level of floating phase Confidence level during the water discharge stage .

[0151] Furthermore, Figure 3 The flowchart of the method for generating independent confidence scores for each floating stage of the kinetic information source provided in the embodiments of this application is as follows: Figure 3 As shown, it includes the following steps:

[0152] Step 310: Extract the vehicle driving force, vehicle motion resistance, and actual acceleration from the dynamic information source;

[0153] Step 320: Calculate the actual acceleration drag based on the vehicle mass and actual acceleration;

[0154] Step 330: Calculate the theoretically expected acceleration resistance based on the vehicle's driving force, vehicle motion resistance, and vehicle dynamics equations;

[0155] Step 340: Based on the actual acceleration drag and the theoretically expected acceleration drag, calculate the difference between acceleration drag and the rate of change of the difference between acceleration drag;

[0156] Step 350: Based on the acceleration resistance difference and the rate of change of the acceleration resistance difference, determine the independent confidence level of the kinetic information source at each floating stage.

[0157] In this embodiment, the independent confidence level of the dynamic information source corresponding to each floating stage is calculated based on vehicle dynamics. The core of this method lies in the fact that the force balance relationship of the vehicle undergoes drastic and characteristic changes during the floating process. By establishing a theoretical model of the vehicle's driving equation and comparing it with the actual measurement values ​​of the sensors in real time, the difference between the floating process and the normal driving process can be calculated. This difference is the key to identifying the floating stage.

[0158] For example, Figure 4 This is a schematic diagram of the longitudinal force on a vehicle provided in an embodiment of this application, such as... Figure 4 As shown, the driving force of the whole vehicle Its main function is to overcome rolling resistance air resistance Slope resistance Accelerating inertial force The force balance equations for the entire vehicle are as follows:

[0159]

[0160] When the vehicle is floating on water, the force balance equation for the entire vehicle changes as follows:

[0161]

[0162] In the above formula, The resistance to a vehicle entering water is a non-linear resistance that only becomes significant when the vehicle is in its buoyancy function, and it is the main cause of the imbalance of forces when the vehicle is wading through water.

[0163] In this embodiment, the acceleration resistance is based on the vehicle's force balance equation. The confidence level based on vehicle dynamics was calculated as the research object.

[0164] Regarding step 320 above, the longitudinal acceleration of the vehicle is measured using an inertial measurement unit (IMU). To calculate the actual acceleration resistance of the vehicle. :

[0165]

[0166] Regarding step 330 above, based on the vehicle's force balance equation during normal driving, the theoretically expected acceleration resistance is calculated. :

[0167]

[0168] For step 340 above, the actual acceleration resistance of the vehicle can be calculated. Compared with theoretically expected acceleration resistance acceleration drag difference :

[0169]

[0170] In the above formula, the difference between acceleration and resistance It directly reflects the degree of abnormality in the vehicle's stress state, such as This means there is additional water resistance. Preventing the vehicle from accelerating.

[0171] Furthermore, after calculating the acceleration-resistance difference... Next, the difference between acceleration and resistance can be analyzed. The difference in acceleration and drag is determined by how it changes over time. The rate of change of the acceleration resistance difference is the rate of change of the acceleration resistance difference.

[0172] In addition, in some implementations, after calculating the acceleration drag difference and the rate of change of the acceleration drag difference, the independent confidence level of the kinetic information source at each floating stage can be calculated through the following judgment steps:

[0173] Step A1: When the rate of change of the acceleration drag difference is greater than zero, select the minimum value from the ratio of the rate of change of the acceleration drag difference to the first slope and the preset confidence upper limit value as the independent confidence level of the dynamic information source in the water entry stage; the first slope is the slope threshold of the rate of change of the acceleration drag difference when it increases.

[0174] Step A2: When the rate of change of the acceleration drag difference is less than or equal to zero, the preset confidence lower limit is used as the independent confidence level of the dynamic information source in the water entry stage;

[0175] Step B: Select the minimum value from the ratio of the acceleration drag difference to the preset force deviation threshold and the preset confidence upper limit value as the independent confidence level of the dynamic information source during the floating phase;

[0176] Step C1: When the rate of change of the acceleration resistance difference is greater than zero, select the minimum value from the ratio of the rate of change of the acceleration resistance difference to the second slope and the preset confidence upper limit value as the independent confidence level of the kinetic information source in the water effluent stage; the second slope is the slope threshold of the rate of change of the acceleration resistance difference when it decreases.

[0177] Step C2: When the rate of change of the acceleration resistance difference is less than or equal to zero, the preset confidence lower limit is used as the independent confidence level of the kinetic information source in the water effluent stage.

[0178] Steps A1 and A2 above are used to calculate the independent confidence level of the dynamic information source during the water entry phase based on vehicle dynamics. Due to the enormous water resistance The sudden appearance and rapid increase of the vehicle's driving force A large portion of it is used to overcome water resistance This leads to a difference in acceleration and resistance. The rate of increase is rapid, therefore the following formula can be used to calculate the independent confidence level of the kinetic information source during the water ingress stage. for:

[0179]

[0180] In the above formula, To accelerate the difference in resistance rate of change, The first slope is used to characterize the difference between acceleration and drag. The slope threshold of the rate of change during rapid increases can be calibrated on a real vehicle, with the upper confidence limit set to 1 and the lower confidence limit set to 0.

[0181] Step B above is used to calculate the independent confidence level of the dynamic information source during the floating phase based on vehicle dynamics. Specifically, the vehicle floats completely in the water thanks to buoyancy. At this point, the tires completely lose traction on the ground, and when the accelerator pedal is pressed, the tires spin freely in the water. Therefore, the force balance model under normal driving conditions completely fails, and the acceleration-drag difference... Since the value remains at an extremely high plateau, the independent confidence level of the dynamic information source during the floating phase can be calculated using the following formula. :

[0182]

[0183] In the above formula, The preset force deviation threshold can be calibrated on a real vehicle, and the preset confidence limit is set to 1.

[0184] Steps C1 and C2 above are used to calculate the independent confidence level of the dynamic information source during the water exit stage based on vehicle dynamics. During the exit phase, the tires gradually regain traction, and water resistance decreases. As the pressure gradually decreased, the vehicle began to switch from floating to normal driving conditions. Since the value begins to decline continuously, the independent confidence level of the kinetic information source during the effluent stage can be calculated using the following formula. :

[0185]

[0186] In the above formula, To accelerate the difference in resistance rate of change, The second slope is used to characterize the difference between acceleration and drag. The threshold for the rapidly decreasing rate of change slope can be calibrated on a real vehicle, with the pre-set upper limit of confidence set to 1.

[0187] In summary, independent confidence levels of the dynamic information source at the three floating stages can be obtained based on vehicle dynamics: , .

[0188] In some embodiments, Figure 4 The flowchart for generating independent confidence levels for each floating stage of the kinematic information source provided in this application embodiment is as follows: Figure 4 As shown, it includes:

[0189] Step 410: Extract the wheel speed difference between the front axle and the rear axle of the vehicle, the front axle slip ratio, and the rear axle slip ratio from the kinematic information source;

[0190] Step 420: Based on wheel speed difference, front axle slip ratio and rear axle slip ratio, determine the independent confidence level of the kinematic information source at each floating stage.

[0191] In this embodiment, the process of the vehicle floating on water will cause uneven force on the front and rear axles of the vehicle and drastic changes in tire adhesion conditions, which will generate two types of characteristic signals at the same time: (1) the wheel speed difference between the front axle and the rear axle of the vehicle. (2) The difference between wheel speed and actual vehicle speed, i.e., the front axle slip ratio of the vehicle. and vehicle rear axle slip ratio .

[0192] Among them, the wheel speed difference between the front axle and the rear axle of the vehicle. The calculation formula is as follows:

[0193]

[0194] In the above formula, For the left front wheel speed, The speed of the right front wheel. For the left rear wheel speed, This refers to the right rear wheel speed, which can be obtained from the vehicle's wheel speed sensors.

[0195] front axle slip ratio of vehicle and vehicle rear axle slip ratio for:

[0196]

[0197]

[0198] In the above formula, This refers to the front axle slip ratio of the vehicle. Rear axle slip ratio; actual vehicle speed This can be obtained through vehicle-mounted positioning sensors. This is the minimum speed among the actual vehicle speeds.

[0199] In some embodiments, based on wheel speed difference, front axle slip ratio, and rear axle slip ratio, the independent confidence level of the kinematic information source at each floating stage can be calculated through the following steps:

[0200] Step 11: Based on the rate of change of wheel speed difference, the lower limit threshold of the rate of change of wheel speed difference, the rate of change of the vehicle's front axle slip ratio, the lower limit threshold of the rate of change of the front axle slip ratio, the preset first weighting coefficient, and the preset upper limit value of confidence, calculate the independent confidence level of the kinematic information source during the water entry phase.

[0201] Step 12: Based on wheel speed difference, wheel speed difference change rate threshold, vehicle front axle slip rate, vehicle rear axle slip rate, front axle slip rate change rate threshold, preset second weighting coefficient and preset confidence upper limit value, calculate the independent confidence of the kinematic information source in the floating phase;

[0202] Step 13: Based on the rate of change of wheel speed difference, the upper limit threshold of the rate of change of wheel speed difference, the rate of change of the vehicle's front axle slip ratio, the upper limit threshold of the rate of change of the front axle slip ratio, the preset third weighting coefficient, and the preset upper limit value of confidence, calculate the independent confidence level of the kinematic information source in the water exit phase.

[0203] For example, the wheel speed difference change rate threshold can be any value between the lower limit threshold and the upper limit threshold of the wheel speed difference change rate; the front axle slip rate change rate threshold can be any value between the lower limit threshold and the upper limit threshold of the front axle slip rate change rate.

[0204] In step 11 above, the independent confidence level of the kinematic information source during the water entry phase is calculated based on vehicle kinematics. During the vehicle's entry into the water, the front axle tires contact the water surface first, while the rear axle tires remain in contact with the ground, resulting in a speed difference between the front and rear axle tires. The slip ratio of the front axle of the vehicle increases sharply. The number of independent confidence levels also increases dramatically, therefore the independent confidence level of the kinetic information source during the water ingress phase can be calculated using the following formula. :

[0205]

[0206] In the above formula, To preset the first weighting coefficient, This is the lower limit threshold for the rate of change of the speed difference between the front and rear axles. The lower limit threshold for the rate of change of front axle slip ratio is set to 1, and the upper limit of the preset confidence level is set to 1.

[0207] In step 12 above, the independent confidence level of the dynamic information source during the floating phase is calculated based on vehicle kinematics. During the floating phase, all tires lose traction, the wheel speed difference between the front and rear axles is small, and the slip ratios of both axles are high. Therefore, the independent confidence level of the dynamic information source during the floating phase can be calculated using the following formula. :

[0208]

[0209] In the above formula, For preset weighting coefficients, The threshold for the rate of change of wheel speed difference. The threshold value for the rate of change of front axle slip ratio is set to 1, with a preset upper limit of confidence.

[0210] In step 13 above, the independent confidence level of the dynamic information source based on vehicle kinematics is calculated during the water exit stage. At that time, during the vehicle's exit from the water phase, the front axle tires made contact with the ground first, while the rear axle tires remained submerged. This resulted in a speed difference between the front and rear axle wheels. The slip ratio of the front axle of the vehicle decreases sharply. The retrieval is also rapid, therefore the independent confidence level of the kinetic information source at the effluent stage can be calculated using the following formula. :

[0211]

[0212] In the above formula, ϵ is a preset third weighting coefficient. This indicates the upper limit threshold for the rate of change of the speed difference between the front and rear axles. This represents the upper limit threshold for the rate of change of front axle slip ratio, with the preset confidence level upper limit set to 1.

[0213] In some embodiments, Figure 5 The flowchart for generating independent confidence levels for each floating stage of the water depth information source provided in this application embodiment is as follows: Figure 5 As shown, it includes:

[0214] Step 510: Extract the wading depth on the left and right sides of the vehicle from the water depth information source;

[0215] Step 520: Based on the wading depth on the left side and the wading depth on the right side of the vehicle, calculate the mean water depth and the rate of change of water depth;

[0216] Step 530: Based on the mean water depth and the rate of change of water depth, determine the independent confidence level of the water depth information source at each floating stage.

[0217] In this embodiment, the wading depth of the vehicle on both sides can be obtained by water depth sensors installed at the rearview mirrors on both sides of the vehicle body. Based on the wading depth of the vehicle body and the rate of change of wading depth, the current floating attitude of the vehicle can be determined, and the independent confidence level corresponding to each floating stage based on the water depth information source can be calculated. , , .

[0218] Among them, the pressure sensors installed on the left and right rearview mirrors respectively measured the wading depth on the left side of the vehicle. , depth of water on the right side of the vehicle Take the average water depth on both sides The overall wading depth of the vehicle:

[0219]

[0220] Rate of change of water depth for:

[0221]

[0222] In some embodiments, based on wheel speed difference, front axle slip ratio, and rear axle slip ratio, the independent confidence level of the kinematic information source at each floating stage can be calculated through the following steps:

[0223] Step 21: Based on the water depth change rate, water depth mean, water depth change rate weighting coefficient during the water entry stage, water depth weighting coefficient during the water entry stage, water depth threshold for vehicle entry, water depth rise rate threshold, and preset confidence upper limit, calculate the independent confidence level of the water depth information source during the water entry stage.

[0224] Step 22: Based on the absolute value of the water depth change rate, the mean water depth, the weighting coefficient of the water depth change rate during the floating phase, the weighting coefficient of the water depth during the floating phase, the allowable rate of water depth fluctuation during the floating phase, the installation height of the vehicle's water depth sensor above the ground, and the preset confidence limit, calculate the independent confidence level of the water depth information source during the floating phase.

[0225] Step 23: Based on the water depth change rate, water depth mean, water depth change rate weighting coefficient during the water discharge stage, water depth weighting coefficient during the water discharge stage, water depth threshold for vehicle water discharge, water depth descent rate threshold, and preset confidence upper limit, calculate the independent confidence level of the water depth information source during the water discharge stage.

[0226] Regarding step 21 above, when calculating the independent confidence level of the water depth information source during the water entry phase based on the mean water depth and the rate of change of water depth, since the front axle of the vehicle enters the water first, the wading depth of the front axle increases rapidly, and the water depth signal collected by the water depth sensor gradually increases. Therefore, the independent confidence level of the water depth information source during the water entry phase can be calculated using the following formula. :

[0227] )

[0228] In the above formula, The threshold water depth for vehicle entry into water. The threshold for the rate of water depth rise. , These are the weighting coefficients for the rate of change of water depth and the water depth during the water entry phase, respectively, satisfying... The preset confidence level is set to 1.

[0229] Regarding step 22 above, when calculating the independent confidence level of the water depth information source during the floating phase based on the mean water depth and the rate of change of water depth, since the vehicle enters the deep water area, both the front and rear axles are submerged, the vehicle body floats, and the wading depth reaches its maximum value with relatively small fluctuations. Therefore, the independent confidence level of the water depth information source during the floating phase can be calculated using the following formula. :

[0230]

[0231] In the above formula, The height above the ground for installing the sensor. For the allowable rate of water depth fluctuation during the floating phase, , These are the weighting coefficients for the rate of change of water depth during the floating phase and the water depth, respectively, satisfying... The preset confidence level is set to 1.

[0232] Regarding step 23 above, when calculating the independent confidence level of the water depth information source during the exit phase based on the mean water depth and the rate of change of water depth, since the vehicle leaves the deep water area, the front axle emerges from the water first, and the wading depth of the front axle decreases rapidly, causing the water depth signal collected by the water depth sensor to gradually decrease. Therefore, the independent confidence level of the water depth information source during the exit phase can be calculated using the following formula. :

[0233]

[0234] In the above formula, The threshold water depth for vehicle exiting the water. The threshold for the rate of water depth descent. , These are the weighting coefficients for the rate of change of water depth and the water depth during the water discharge stage, respectively, satisfying... The preset confidence level is set to 1.

[0235] In some embodiments, Figure 6 The flowchart for generating independent confidence levels for the suspension information source provided in this application embodiment at each floating stage is as follows: Figure 6 As shown, it includes:

[0236] Step 610: Extract the front axle suspension holding force, the rate of change of the front axle suspension holding force, the rear axle suspension holding force, and the rate of change of the rear axle suspension holding force from the suspension information source;

[0237] Step 620: Based on the vehicle's front axle suspension holding force, the rate of change of the front axle suspension holding force, the vehicle's rear axle suspension holding force, and the rate of change of the rear axle suspension holding force, determine the independent confidence level of the suspension information source for each floating stage.

[0238] In this embodiment, as the vehicle gradually enters the water, the buoyancy it experiences gradually increases, resulting in a corresponding decrease in the vertical load acting on the tires and suspension. When the vehicle is completely floating, the buoyancy it experiences is equal to its own weight, and the suspension system no longer bears the weight of the vehicle body, so the suspension holding force approaches zero. When the vehicle emerges from the water, the suspension system gradually experiences force, and the suspension holding force gradually recovers.

[0239] In this embodiment, the confidence level for the three floating stages is calculated by monitoring the absolute value and rate of change or trend of the suspension holding force. , , .

[0240] Among them, the holding force of the left front axle suspension can be obtained by force sensors installed on each suspension. Right front axle suspension holding force Left rear axle suspension holding force Right rear axle suspension holding force .

[0241] Among them, the front axle suspension holding force of the vehicle Rear axle suspension holding force They are respectively:

[0242]

[0243]

[0244] The rate of change of the front and rear axle suspension holding force can be obtained. They are respectively:

[0245]

[0246]

[0247] In some embodiments, based on the vehicle's front axle suspension holding force, the rate of change of the front axle suspension holding force, the vehicle's rear axle suspension holding force, and the rate of change of the rear axle suspension holding force, the independent confidence level of the suspension holding force information source at each floating stage can be calculated through the following steps:

[0248] Step 31: Based on the vehicle's front axle suspension holding force, front axle suspension holding force change rate, front axle suspension holding force decrease rate threshold, front axle suspension holding force change rate weighting coefficient and front axle suspension holding force weighting coefficient during the water entry stage, the front axle suspension holding force when the vehicle is stationary on a level road surface, and the preset confidence upper limit value, calculate the independent confidence level of the suspension information source corresponding to the water entry stage.

[0249] Step 32: Based on the vehicle's front axle suspension holding force, vehicle's rear axle suspension holding force, the front axle suspension holding force when the vehicle is stationary on a level road, the rear axle suspension holding force when the vehicle is stationary on a level road, the weight coefficient of the front axle suspension holding force during the floating phase, the weight coefficient of the rear axle suspension holding force during the floating phase, and the preset confidence upper limit, calculate the independent confidence level of the suspension information source during the floating phase.

[0250] Step 33: Based on the front axle suspension holding force change rate, the vehicle's front axle suspension holding force, the front axle suspension holding force when the vehicle is stationary on a level road, the front axle suspension holding force rise rate threshold, the weighting coefficient of the front axle suspension holding force change rate during the water exit stage, the weighting coefficient of the front axle suspension holding force during the water exit stage, and the preset confidence upper limit, calculate the independent confidence level of the suspension information source during the water exit stage.

[0251] Regarding step 31 above, the independent confidence level of the suspension information source during the water entry phase is calculated based on the suspension holding force. During the water immersion phase, the front axle suspension holding force rapidly decreases from its normal value. Therefore, the independent confidence level of the suspension information source during the water immersion phase can be calculated using the following formula. :

[0252]

[0253] In the above formula, The threshold value for the rate of decrease in front axle suspension holding force. This refers to the suspension holding force on the front axle when the vehicle is stationary on a level road surface. and The weighting coefficients for the rate of change of front axle suspension holding force and the front axle suspension holding force during the water entry phase satisfy the following conditions: The preset confidence level is set to 1.

[0254] Regarding step 32 above, the independent confidence level of the suspension information source during the floating phase is calculated. During the floating phase, compared to normal driving conditions, the suspension holding forces of both the front and rear axles decrease significantly and stabilize at an extremely low level (close to zero). Therefore, the independent confidence level of the suspension information source during the floating phase can be calculated using the following formula. :

[0255]

[0256] In the above formula, , These are the front axle suspension holding force and the rear axle suspension holding force when the vehicle is stationary on a level road surface, respectively. and The weighting coefficients for the front and rear axle suspension holding forces satisfy... The preset confidence level is set to 1.

[0257] Regarding step 33 above, the independent confidence level of the suspension information source during the water discharge phase is calculated. During the water exit phase, the front axle suspension holding force rapidly increases from a low value and gradually returns to the load level of normal driving. Therefore, the independent confidence level of the suspension information source during the water exit phase can be calculated using the following formula. :

[0258]

[0259] In the above formula, The threshold value for the rate of increase of the front axle suspension holding force. and The weighting coefficients for the rate of change of front axle suspension holding force and the front axle suspension holding force during the water exit phase satisfy the following conditions: The preset confidence level is set to 1.

[0260] In some embodiments, Figure 7 The flowchart for generating independent confidence scores for each floating stage of the foreground image information source provided in this application embodiment is as follows: Figure 7 As shown, it includes:

[0261] Step 710: Extract the image of the vehicle's front from the front image information source;

[0262] Step 720: Obtain the water surface area and road surface area in the image in front of the vehicle;

[0263] Step 730: Based on the water surface area and the road surface area, calculate the proportion of the water-involved area and the rate of change of the proportion of the water-involved area;

[0264] Step 740: Based on the proportion of the water-touched area and the rate of change of the proportion of the water-touched area, determine the independent confidence level of the image information source in front at each floating stage.

[0265] In this embodiment, the rate of change in the relative proportion of the water surface area to the road surface area in the field of view of the forward-facing camera installed at the front of the vehicle during the floating process directly reflects the relative motion relationship between the vehicle body and the water level.

[0266] For example, Figure 8 This is a schematic diagram illustrating the changes in images acquired during the vehicle's entry into water phase, as provided in an embodiment of this application. Figure 8 As shown, the area of ​​the wading zone in the image ahead of the vehicle is continuously increasing. Furthermore, Figure 9 This is a schematic diagram illustrating the changes in images acquired during the vehicle's water exit phase, as provided in the embodiments of this application. Figure 9 As shown, the area of ​​the wading zone in the image in front of the vehicle is continuously decreasing.

[0267] In this embodiment, the independent confidence level of the forward image information source at the three floating stages can be calculated by monitoring the proportion of the wading area in the image ahead of the vehicle and the rate of change of the proportion of the wading area.

[0268] First, image signals are acquired using the vehicle's front-view camera to obtain video stream frames. Each acquired frame is then segmented pixel-level based on image semantic segmentation to obtain pixel classifications including water surface and road surface. Finally, the proportion of the wading area in the image is calculated based on the number of water surface pixels in each frame. :

[0269]

[0270] In the above formula, These are the pixels in the current image that are classified as water. This is the total number of pixels in the current image.

[0271] Furthermore, to represent the rate of increase or decrease in the area of ​​water in the image, the rate of change of the proportion of water-bound areas in the image is calculated. for:

[0272]

[0273] In some embodiments, based on the proportion of the wading area and the rate of change of the proportion of the wading area, the independent confidence level of the foreground image information source at each floating stage can be calculated by the following steps:

[0274] Step 41: Based on the rate of change of the proportion of the wading area, the proportion of the wading area, the weight of the rate of change of the proportion of the wading area in the front image of the vehicle during the water entry stage, the threshold for the increase of the rate of change of the proportion of the wading area, the weight of the proportion of the wading area, the saturation threshold of the proportion of the wading area, and the preset confidence upper limit, calculate the independent confidence of the front image information source during the water entry stage.

[0275] Step 42: Based on the proportion of wading area, the weight of the proportion of wading area in the image in front of the vehicle during the floating stage, the high proportion threshold of wading area, and the preset confidence upper limit, calculate the independent confidence of the image information source in front during the floating stage.

[0276] Step 43: Based on the rate of change of the proportion of the wading area, the proportion of the wading area, the weight of the rate of change of the proportion of the wading area in the image in front of the vehicle during the water exit stage, the threshold for the decrease of the rate of change of the proportion of the wading area, the weight of the proportion of the wading area, the threshold for the low proportion of the wading area, and the preset confidence upper limit, calculate the independent confidence of the image information source in front of the vehicle during the water exit stage.

[0277] Regarding step 41 above, during the water entry phase, as the vehicle moves into deeper water, the water level rises rapidly, and the road surface area decreases rapidly, resulting in a sharp increase in the proportion of water in the image. The rate of change is a very high positive value. Therefore, the independent confidence level of the preceding image information source during the water entry phase can be calculated using the following formula:

[0278]

[0279] In the above formula, The weight of the rate of change of the proportion of the wading area in the image during the water entry phase. The threshold for the rate of increase in the percentage of water-bound area in the image during the water entry phase. The weights represent the proportion of water-bound areas in the images during the water entry phase. The saturation threshold for the proportion of water-affected areas is set, with a pre-set upper limit of confidence level of 1.

[0280] Regarding step 42 above, during the floating phase, the vehicle is suspended in the water, and the road surface area disappears, resulting in a consistently high proportion of water in the image. Therefore, the independent confidence level of the foreground image source during the floating phase can be calculated using the following formula. :

[0281]

[0282] In the above formula, The weights represent the proportion of wading areas in the floating phase image. The threshold for high proportion of water-related areas is set with a pre-set upper limit of 1 for confidence level.

[0283] Regarding step 43 above, during the water exit phase, the vehicle leaves the deep water area, the water level relatively drops, and the road surface reappears, causing a sharp decrease in the proportion of the water area in the image, with a very high negative rate of change. Therefore, the independent confidence level of the foreground image information source during the water exit phase can be calculated using the following formula. :

[0284]

[0285] In the above formula, The weight of the rate of change of the proportion of water-covered area in the image during the water discharge phase. The threshold for the rate of decrease in the percentage of water-covered area in the image during the water discharge phase. The weighting of the proportion of water-covered areas in the images during the water discharge phase. The threshold for low water-related areas is set, with a pre-set upper limit of confidence level of 1.

[0286] In summary, the independent confidence levels of the five information sources at each floating stage were obtained, namely the confidence levels for the water entry stage. Confidence level of floating phase Confidence level during the water discharge stage This is used to further achieve the aforementioned confidence fusion.

[0287] in, They represent: 1-Dynamic information source, 2-Kinematic information source, 3-Water depth information source, 4-Suspension information source, and 5-Forward image information source, respectively.

[0288] Through the above embodiments, by tailoring a dedicated feature extraction algorithm for each sensor to calculate the independent confidence level of each information source at each floating stage, and then combining the multi-source heterogeneous sensor information fusion architecture, the independent confidence levels are fused to obtain the fused confidence level. This enables continuous and accurate identification of which stage the vehicle is in, namely the water exit stage, the floating stage, or the water exit stage, providing a more refined decision-making basis for vehicle control.

[0289] Furthermore, traditional single-sensor judgments may fail or become unreliable due to their own limitations, external interference, or extreme scenarios. This solution, however, calculates independent confidence signals from five information sources at each buoyancy stage using the aforementioned embodiments, and then fuses the confidence scores from the five different information sources. By leveraging their complementarity and redundancy, it ultimately outputs a more reliable, accurate, and robust buoyancy stage judgment.

[0290] Figure 10 This is a schematic diagram of the vehicle floating stage identification framework provided in the embodiments of this application, as shown below. Figure 10 As shown, its core consists of a multi-source heterogeneous sensor perception layer and a central decision fusion layer based on DS evidence theory.

[0291] The perception layer integrates five heterogeneous sensors—vehicle dynamics, vehicle kinematics, water depth, suspension holding force, and visual images—to construct a redundant physical information acquisition network. This network can accurately capture characteristic signals from different dimensions during the three stages of entering, floating, and exiting the water, and calculate the independent confidence level of each information source for each floating stage. Specifically, it includes the following five groups:

[0292] (1) The independent confidence levels of the dynamic information sources obtained from vehicle dynamics calculations at each floating stage are respectively , ;

[0293] (2) The independent confidence levels of the kinematic information sources obtained from vehicle kinematics calculations at each floating stage are as follows: ;

[0294] (3) The independent confidence levels of the water depth information source calculated based on the water depth signal at each floating stage are as follows: , , ;

[0295] (4) The independent confidence levels of the suspension information source calculated based on the suspension holding force at each floating stage are as follows: , , ;

[0296] (5) The independent confidence levels of the foreground image information source calculated based on the image signal at each floating stage are as follows: , , .

[0297] After calculating the independent confidence levels, the central fusion layer uses DS evidence theory to transform the independent confidence levels output by each sensor into probability quality allocations. It resolves sensor contradictions by calculating the conflict coefficient between evidences and strengthens the propositional reliability of collaborative support.

[0298] The vehicle floating stage identification framework provided in this application not only achieves dynamic and accurate identification of the floating stage, but also significantly improves the system's decision-making robustness and safety in complex water environments by explicitly managing uncertainty.

[0299] In some embodiments, after identifying the floating stage of the vehicle, sensor signals collected by each of the vehicle's sensors can be further acquired; then, based on at least one of the water depth signal, vehicle suspension holding force signal, vehicle front and rear axle wheel speed difference signal, and front and rear axle wheel speed change signal, the degree of water penetration of the vehicle's front axle is determined, and the degree of water penetration is divided into front axle not in water, front axle slightly in water, and front axle deeply in water; and based at least on the degree of water penetration of the vehicle's front axle and the vehicle pitch angle signal, the floating attitude of the vehicle during the floating stage is identified.

[0300] The sensor signals include water depth signals, vehicle suspension holding force signals, vehicle front and rear axle wheel speed difference signals, front and rear axle wheel speed change signals, and vehicle pitch angle signals.

[0301] Figure 11 This is a schematic diagram of the vehicle's floating posture provided in the embodiments of this application, such as... Figure 11 As shown, the vehicle can be divided into 13 attitudes from the moment it enters the water to the moment it exits the water.

[0302] In this embodiment, a multi-source information fusion process for vehicle buoyancy recognition can be implemented based on a full-process state machine. The inputs to the full-process state machine include water depth signal, suspension holding force signal, vehicle pitch angle signal, front and rear axle wheel speed difference signal, and front and rear axle wheel speed change rate signal. Combined with... Figure 11 The floating attitude output by the full-process state machine includes:

[0303] (1) The front axle of the vehicle is not in the water and the vehicle is not pitching. The vehicle is identified as being in attitude 1, i.e., not in the water attitude.

[0304] (2) When the front axle of the vehicle is not submerged in water, but the vehicle pitches (the front of the vehicle is submerged), the vehicle is identified as being in attitude 2, and the entire process state machine switches to attitude 2, i.e., pre-submersion attitude.

[0305] (3) When the front axle of the vehicle is slightly submerged in water, the vehicle pitches (the front of the vehicle is submerged), the vehicle is identified as being in attitude 3, and the entire process state machine switches to attitude 3, i.e. shallow water submersion attitude.

[0306] (4) When the front axle of the vehicle is submerged in water but the rear axle is not, the front axle suspension holding force drops to 0, the front axle wheel speed increases sharply and the difference between the front and rear axle wheel speeds increases, the vehicle pitches (the front of the vehicle is submerged), the vehicle is identified as being in posture 4, and the entire process state machine switches to posture 4, i.e., the deep submersion posture.

[0307] (5) When the front axle of the vehicle is submerged in water and the rear axle is slightly submerged, the front axle suspension holding force drops sharply and the vehicle pitches (the front of the vehicle is submerged). The vehicle is identified as being in attitude 5, and the entire process state machine switches to attitude 5, i.e., the pre-floating attitude.

[0308] (6) When the front axle of the vehicle is submerged in water, the holding force of the rear axle suspension drops sharply, the holding force of the front and rear axles of the vehicle is almost zero, the rear axle wheel speed increases sharply, the vehicle pitches (the front of the vehicle is submerged), the vehicle is identified as being in attitude 6, and the entire process state machine switches to attitude 6, i.e., the submerged floating attitude.

[0309] (7) When the front axle of the vehicle is submerged in water, the holding force of the front and rear axle suspensions is almost zero, the vehicle attitude is stable and there is no pitching, the vehicle is identified as being in attitude 7, and the entire process state machine switches to attitude 7, that is, the horizontal floating attitude.

[0310] (8) When the front axle of the vehicle is submerged in water, the holding force of the front and rear axle suspensions is almost zero, the vehicle pitches (the front of the vehicle lifts up), the front axle wheel speed drops sharply, the vehicle is identified as being in posture 8, and the entire process state machine switches to posture 8, i.e., the head-lifting floating posture.

[0311] (9) When the front axle of the vehicle is slightly submerged in water, the vehicle pitches (the front of the vehicle lifts up), the rear axle suspension holding force is almost zero, the front axle wheel speed decreases sharply and the speed difference between the front and rear axles decreases, the vehicle is identified as being in posture 9, and the entire process state machine switches to posture 9, that is, the front wheels are on the ground and the rear wheels are floating.

[0312] (10) When the front axle of the vehicle is not in the water, the vehicle pitches (the front of the vehicle lifts up more), the rear axle suspension holding force is almost zero, the vehicle is identified as being in attitude 10, and the whole process state machine switches to attitude 10, that is, the front wheel out of the water attitude.

[0313] (11) When the front axle of the vehicle is not in the water, the vehicle pitches (the front of the vehicle lifts up further), the rear axle suspension holding force increases, the rear axle wheel speed drops sharply, the vehicle is identified as being in posture 11, and the whole process state machine switches to posture 11, that is, the rear wheel is on the ground and the front wheel is out of the water posture.

[0314] (12) When the front axle of the vehicle is not submerged in water, the vehicle pitches (the degree of nose lift of the front of the vehicle is relieved), the rear axle suspension holding force of the vehicle increases further, the vehicle is identified as being in attitude 12, and the whole process state machine switches to attitude 12, that is, the fully submerged attitude.

[0315] (13) When the front axle of the vehicle is not submerged in water, the vehicle attitude is stable and there is no pitching, the suspension holding force of the front and rear axles of the vehicle is restored to the value when driving normally on land, the vehicle is identified as being in attitude 13, and the entire process state machine is switched to attitude 13, which is the normal driving attitude.

[0316] This embodiment synchronously collects, preprocesses, and correlates water depth signals, vehicle suspension holding force signals, vehicle front and rear axle wheel speed difference signals, front and rear axle wheel speed change signals, and vehicle pitch angle signals. It unifies environmental changes, support changes, wheel system motion changes, and vehicle body attitude changes during the vehicle's wading process into a single recognition framework. This enables more accurate differentiation between the water entry stage, floating stage, and water exit stage, and further identifies the specific floating attitude within each stage. This alleviates the problems of misjudgment, missed judgment, and stage discontinuity caused by relying on a single threshold or fixed logic in traditional technologies, and improves the accuracy, robustness, and real-time control applicability of vehicle floating attitude recognition under complex wading conditions.

[0317] Figure 12 This is a schematic diagram of the vehicle floating stage identification device provided in this application, as shown below. Figure 12 As shown, the vehicle floating stage identification device 1200 provided in this embodiment includes:

[0318] The construction module 1210 is used to construct at least two information sources based on the data collected by various sensors of the vehicle. The at least two information sources include at least two of the following: vehicle dynamics information source, kinematics information source, water depth information source, suspension information source, and forward image information source.

[0319] The extraction module 1220 is used to extract features from at least two information sources for each floating stage, and generate an independent confidence level for each information source in each floating stage. The independent confidence level of any information source in the floating stage is used to characterize the probability that the vehicle is in the floating stage as determined by the information source.

[0320] The generation module 1230 is used to merge the independent confidence scores of various information sources at the same floating stage to generate a comprehensive confidence score for each floating stage.

[0321] The identification module 1240 is used to identify the current floating stage of the vehicle based on the comprehensive confidence level of each floating stage. The floating stage includes the entry stage, the floating stage, or the exit stage.

[0322] In one possible implementation, the identification module can be specifically used to: determine the target comprehensive confidence level from various comprehensive confidence levels; and determine the floating stage corresponding to the target comprehensive confidence level as the floating stage of the vehicle at the current moment, wherein the target comprehensive confidence level is the one with the largest value among the various comprehensive confidence levels.

[0323] In one possible implementation, the generation module can specifically be used for:

[0324] Obtain the uncertainty ratio coefficient corresponding to each information source;

[0325] Based on the uncertainty ratio coefficient corresponding to each information source, the independent confidence level of each information source in each floating stage, and the sum of independent confidence levels of each information source in each floating stage, the basic probability distribution of each information source for each floating stage is calculated.

[0326] Based on the DS evidence theory, the basic probability distribution of each information source for each floating stage is probabilistically fused to generate a comprehensive confidence level for each floating stage.

[0327] In one possible implementation, the extraction module is specifically used for:

[0328] If at least two information sources include a dynamic information source, then the vehicle driving force, vehicle motion resistance, and actual acceleration are extracted from the dynamic information source.

[0329] Calculate the actual acceleration drag based on the vehicle mass and actual acceleration.

[0330] Based on the vehicle's driving force, vehicle motion resistance, and vehicle dynamics equations, the theoretically expected acceleration resistance is calculated.

[0331] Based on the actual acceleration drag and the theoretically expected acceleration drag, calculate the difference in acceleration drag and the rate of change of the difference in acceleration drag;

[0332] Based on the acceleration resistance difference and the rate of change of the acceleration resistance difference, the independent confidence level of the kinetic information source is determined for each floating stage.

[0333] In one possible implementation, the extraction module is specifically used for:

[0334] When the rate of change of the acceleration drag difference is greater than zero, the minimum value is selected from the ratio of the rate of change of the acceleration drag difference to the first slope and the preset confidence upper limit value, which is used as the independent confidence level of the dynamic information source in the water entry stage; the first slope is the slope threshold of the rate of change of the acceleration drag difference when it increases.

[0335] When the rate of change of acceleration drag difference is less than or equal to zero, the preset confidence lower limit is used as the independent confidence level of the dynamic information source in the water entry stage.

[0336] The minimum value is selected from the ratio of the acceleration drag difference to the preset force deviation threshold and the preset confidence upper limit value as the independent confidence level of the dynamic information source during the floating phase.

[0337] When the rate of change of the acceleration resistance difference is greater than zero, the minimum value is selected from the ratio of the rate of change of the acceleration resistance difference to the second slope and the preset confidence upper limit value, which is used as the independent confidence level of the kinetic information source in the water discharge stage; the second slope is the slope threshold of the rate of change of the acceleration resistance difference when it decreases.

[0338] When the rate of change of acceleration resistance difference is less than or equal to zero, the preset confidence lower limit is used as the independent confidence level of the kinetic information source in the water discharge stage.

[0339] In one possible implementation, the extraction module is specifically used to: extract the wheel speed difference between the front axle and the rear axle of the vehicle, the front axle slip ratio, and the rear axle slip ratio from the kinematic information source if at least two information sources include a kinematic information source; and determine the independent confidence level of the kinematic information source at each floating stage based on the wheel speed difference, the front axle slip ratio, and the rear axle slip ratio.

[0340] In one possible implementation, the extraction module is specifically used for:

[0341] Based on the rate of change of wheel speed difference, the lower limit threshold of the rate of change of wheel speed difference, the rate of change of the vehicle's front axle slip ratio, the lower limit threshold of the rate of change of the front axle slip ratio, the preset first weight coefficient, and the preset upper limit value of confidence, the independent confidence of the kinematic information source during the water entry phase is calculated.

[0342] Based on wheel speed difference, wheel speed difference change rate threshold, vehicle front axle slip rate, vehicle rear axle slip rate, front axle slip rate change rate threshold, preset second weight coefficient and preset confidence upper limit value, calculate the independent confidence of the kinematic information source in the floating phase;

[0343] Based on the rate of change of wheel speed difference, the upper limit threshold of the rate of change of wheel speed difference, the rate of change of the vehicle's front axle slip ratio, the upper limit threshold of the rate of change of the front axle slip ratio, the preset third weighting coefficient, and the preset upper limit value of confidence, the independent confidence level of the kinematic information source during the water exit phase is calculated.

[0344] In one possible implementation, the extraction module is specifically used for:

[0345] If at least two information sources include a water depth information source, then extract the wading depth on the left side of the vehicle and the wading depth on the right side of the vehicle from the water depth information source.

[0346] Calculate the mean water depth and the rate of change of water depth based on the wading depth on the left and right sides of the vehicle.

[0347] Based on the mean water depth and the rate of change of water depth, the independent confidence level of the water depth information source is determined for each floating stage.

[0348] In one possible implementation, the extraction module is specifically used for:

[0349] Based on the water depth change rate, water depth mean, water depth change rate weighting coefficient during the water entry stage, water depth weighting coefficient during the water entry stage, water depth threshold for vehicle water entry, water depth rise rate threshold, and preset confidence upper limit, calculate the independent confidence level of the water depth information source during the water entry stage.

[0350] Based on the absolute value of the rate of change of water depth, the mean water depth, the weighting coefficient of the rate of change of water depth during the floating phase, the weighting coefficient of water depth during the floating phase, the allowable rate of water depth fluctuation during the floating phase, the installation height of the vehicle's water depth sensor above the ground, and the preset upper limit of confidence, the independent confidence level of the water depth information source during the floating phase is calculated.

[0351] Based on the water depth change rate, water depth mean, water depth change rate weighting coefficient during the water exit stage, water depth weighting coefficient during the water exit stage, water depth threshold for vehicle water exit, water depth descent rate threshold, and a preset confidence upper limit, the independent confidence level of the water depth information source during the water exit stage is calculated.

[0352] In one possible implementation, the extraction module is specifically used to: if at least two information sources include a suspension information source, extract the vehicle's front axle suspension holding force, the rate of change of the front axle suspension holding force, the vehicle's rear axle suspension holding force, and the rate of change of the rear axle suspension holding force from the suspension information source; and determine the independent confidence level of the suspension information source corresponding to each floating stage based on the vehicle's front axle suspension holding force, the rate of change of the front axle suspension holding force, the vehicle's rear axle suspension holding force, and the rate of change of the rear axle suspension holding force.

[0353] In one possible implementation, the extraction module is specifically used for:

[0354] Based on the vehicle's front axle suspension holding force, front axle suspension holding force change rate, front axle suspension holding force decrease rate threshold, front axle suspension holding force change rate weighting coefficient and front axle suspension holding force weighting coefficient during the water entry stage, the front axle suspension holding force when the vehicle is stationary on a level road surface, and the preset confidence upper limit value, calculate the independent confidence level of the suspension information source during the water entry stage.

[0355] Based on the vehicle's front axle suspension holding force, vehicle's rear axle suspension holding force, the front axle suspension holding force when the vehicle is stationary on a level road, the rear axle suspension holding force when the vehicle is stationary on a level road, the weight coefficient of the front axle suspension holding force during the floating phase, the weight coefficient of the rear axle suspension holding force during the floating phase, and the preset confidence upper limit, calculate the independent confidence level of the suspension information source during the floating phase.

[0356] Based on the front axle suspension holding force change rate, the vehicle's front axle suspension holding force, the front axle suspension holding force when the vehicle is stationary on a level road, the front axle suspension holding force rise rate threshold, the weighting coefficient of the front axle suspension holding force change rate during the water exit stage, the weighting coefficient of the front axle suspension holding force during the water exit stage, and the preset confidence upper limit, the independent confidence level of the suspension information source during the water exit stage is calculated.

[0357] In one possible implementation, the extraction module is specifically used for:

[0358] If at least two information sources include a frontal image information source, then extract the frontal image of the vehicle from the frontal image information source;

[0359] Acquire the water and road surface areas in the image in front of the vehicle;

[0360] Based on the water surface area and the road surface area, calculate the proportion of water-related areas and the rate of change of the proportion of water-related areas;

[0361] Based on the proportion of water-covered areas and the rate of change of the proportion of water-covered areas, the independent confidence level of the image information source at each floating stage is determined.

[0362] In one possible implementation, the extraction module is specifically used for:

[0363] Based on the rate of change of the proportion of the wading area, the proportion of the wading area, the weight of the rate of change of the proportion of the wading area in the front image of the vehicle during the water entry stage, the threshold for the increase of the rate of change of the proportion of the wading area, the weight of the proportion of the wading area, the saturation threshold of the proportion of the wading area and the preset confidence upper limit, the independent confidence of the front image information source during the water entry stage is calculated.

[0364] Based on the proportion of wading area, the weight of the proportion of wading area in the image ahead of the vehicle during the floating stage, the high proportion threshold of wading area, and the preset confidence upper limit, the independent confidence of the image information source ahead during the floating stage is calculated.

[0365] Based on the rate of change of the proportion of the wading area, the proportion of the wading area, the weight of the rate of change of the proportion of the wading area in the image in front of the vehicle during the water exit stage, the threshold for the decrease of the rate of change of the proportion of the wading area, the weight of the proportion of the wading area, the threshold for the low proportion of the wading area, and the preset confidence upper limit, the independent confidence of the image information source in front during the water exit stage is calculated.

[0366] In one possible implementation, the vehicle floating phase identification device further includes an attitude recognition module for:

[0367] The sensor signals collected by each of the vehicle's sensors are acquired. These sensor signals include water depth signal, vehicle suspension holding force signal, vehicle front and rear axle wheel speed difference signal, front and rear axle wheel speed change signal, and vehicle pitch angle signal.

[0368] Based on at least one of the following signals: water depth signal, vehicle suspension holding force signal, vehicle front and rear axle wheel speed difference signal, and front and rear axle wheel speed change signal, the degree of water penetration of the vehicle's front axle is determined. The degree of water penetration is divided into three categories: front axle not submerged, front axle slightly submerged, and front axle deeply submerged.

[0369] The floating attitude of the vehicle during the floating phase can be identified based at least on the degree of water entry of the front axle and the vehicle pitch angle signal.

[0370] The vehicle floating stage identification device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0371] Figure 13 A schematic diagram of the structure of the electronic device provided in this application. Figure 13 As shown, the electronic device 1300 provided in this embodiment includes at least one processor 1301 and a memory 1302. Optionally, the electronic device 1300 further includes a communication component 1303. The processor 1301, memory 1302, and communication component 1303 are connected via a bus.

[0372] In a specific implementation, at least one processor 1301 executes computer execution instructions stored in memory 1302, causing at least one processor 1301 to perform the above-described method.

[0373] The specific implementation process of processor 1301 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0374] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0375] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0376] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0377] This application also provides a vehicle, including a vehicle body and the aforementioned electronic device disposed in the vehicle body, the electronic device being capable of performing the aforementioned method steps.

[0378] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0379] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0380] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0381] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0382] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0383] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0384] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0385] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0386] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for identifying the floating stage of a vehicle, characterized in that, include: Based on the data collected by various sensors of the vehicle, at least two information sources are constructed, including at least two of the following: vehicle dynamics information source, kinematics information source, water depth information source, suspension information source, and forward image information source. For each floating stage, feature extraction is performed on the at least two information sources to generate an independent confidence level for each information source in each floating stage. The independent confidence level of any information source in the floating stage is used to characterize the probability that the vehicle is in the floating stage as determined by the information source. The independent confidence scores of each information source at the same floating stage are merged to generate a comprehensive confidence score for each floating stage. Based on the overall confidence level of each floating stage, the floating stage of the vehicle at the current moment is identified, and the floating stage includes the entry stage, the floating stage, or the exit stage.

2. The method according to claim 1, characterized in that, The process of identifying the current floating stage of the vehicle based on the comprehensive confidence level of each floating stage includes: The target comprehensive confidence score is determined from the various comprehensive confidence scores, wherein the target comprehensive confidence score is the largest value among the various comprehensive confidence scores; The floating stage corresponding to the target comprehensive confidence level is determined as the floating stage of the vehicle at the current moment.

3. The method according to claim 1, characterized in that, The process of fusing the independent confidence levels of various information sources at the same floating stage to generate a comprehensive confidence level for each floating stage includes: Obtain the uncertainty ratio coefficient corresponding to each information source; Based on the uncertainty ratio coefficient corresponding to each information source, the independent confidence level of each information source in each floating stage, and the sum of independent confidence levels of each information source in each floating stage, the basic probability distribution of each information source for each floating stage is calculated. Based on the DS evidence theory, the basic probability distribution of each information source for each floating stage is probabilistically fused to generate a comprehensive confidence level for each floating stage.

4. The method according to any one of claims 1-3, characterized in that, The step of extracting features from the at least two information sources and generating an independent confidence level for each information source at each floating stage includes: If the dynamic information source is included among the at least two information sources, then the vehicle driving force, vehicle motion resistance and actual acceleration are extracted from the dynamic information source. Calculate the actual acceleration drag based on the vehicle mass and the actual acceleration; Based on the vehicle driving force, the vehicle motion resistance, and the vehicle dynamics equations, calculate the theoretical expected acceleration resistance; Based on the actual acceleration drag and the theoretical expected acceleration drag, calculate the acceleration drag difference and the rate of change of the acceleration drag difference; Based on the acceleration resistance difference and the rate of change of the acceleration resistance difference, the independent confidence level of the kinetic information source is determined for each floating stage.

5. The method according to claim 4, characterized in that, The determination of the independent confidence level of the kinetic information source at each floating stage based on the acceleration resistance difference and the rate of change of the acceleration resistance difference includes: When the rate of change of the acceleration resistance difference is greater than zero, the minimum value is selected from the ratio of the rate of change of the acceleration resistance difference to the first slope and the preset confidence upper limit value, and is used as the independent confidence level of the dynamic information source in the water entry stage; the first slope is the slope threshold of the rate of change of the acceleration resistance difference when it increases; When the rate of change of the acceleration resistance difference is less than or equal to zero, the preset confidence lower limit value will be used as the independent confidence level of the dynamic information source in the water entry stage. The minimum value is selected from the ratio of the acceleration resistance difference to the preset force deviation threshold and the preset confidence upper limit value, and is used as the independent confidence level of the dynamic information source in the floating stage. When the rate of change of the acceleration resistance difference is greater than zero, the minimum value is selected from the ratio of the rate of change of the acceleration resistance difference to the second slope and the preset confidence upper limit value, and is used as the independent confidence level of the dynamic information source in the water effluent stage; the second slope is the slope threshold of the rate of change of the acceleration resistance difference when it decreases. When the rate of change of the acceleration resistance difference is less than or equal to zero, the preset confidence lower limit is used as the independent confidence level of the dynamic information source in the water discharge stage.

6. The method according to any one of claims 1-3, characterized in that, The step of extracting features from the at least two information sources and generating an independent confidence level for each information source at each floating stage includes: If the kinematic information source is included among the at least two information sources, then the wheel speed difference between the front axle and the rear axle of the vehicle, the front axle slip ratio, and the rear axle slip ratio of the vehicle are extracted from the kinematic information source. Based on the wheel speed difference, the vehicle's front axle slip ratio, and the vehicle's rear axle slip ratio, the independent confidence level of the kinematic information source is determined for each floating stage.

7. The method according to claim 6, characterized in that, The determination of the independent confidence level of the kinematic information source for each floating stage based on the wheel speed difference, the front axle slip ratio, and the rear axle slip ratio includes: Based on the rate of change of wheel speed difference, the lower limit threshold of the rate of change of wheel speed difference, the rate of change of the vehicle front axle slip ratio, the lower limit threshold of the rate of change of the front axle slip ratio, the preset first weighting coefficient, and the preset upper limit of confidence, the independent confidence level of the kinematic information source during the water entry phase is calculated. Based on the wheel speed difference, the wheel speed difference change rate threshold, the vehicle front axle slip rate, the vehicle rear axle slip rate, the front axle slip rate change rate threshold, the preset second weighting coefficient, and the preset confidence upper limit value, calculate the independent confidence level of the kinematic information source during the floating phase; Based on the rate of change of wheel speed difference, the upper limit threshold of the rate of change of wheel speed difference, the rate of change of the vehicle's front axle slip ratio, the upper limit threshold of the rate of change of the front axle slip ratio, the preset third weighting coefficient, and the preset confidence upper limit value, the independent confidence level of the kinematic information source during the water exit phase is calculated.

8. The method according to any one of claims 1-3, characterized in that, The step of extracting features from the at least two information sources and generating an independent confidence level for each information source at each floating stage includes: If the water depth information source is included among the at least two information sources, then the wading depth on the left side of the vehicle and the wading depth on the right side of the vehicle are extracted from the water depth information source. Based on the wading depths on the left and right sides of the vehicle, calculate the mean water depth and the rate of change of water depth. Based on the mean water depth and the rate of change of water depth, the independent confidence level of the water depth information source is determined for each floating stage.

9. The method according to claim 8, characterized in that, The determination of the independent confidence level of the water depth information source at each floating stage based on the mean water depth and the rate of change of water depth includes: Based on the water depth change rate, the water depth mean, the weighting coefficient of the water depth change rate during the water entry stage, the water depth weighting coefficient during the water entry stage, the water depth threshold for vehicle water entry, the water depth rise rate threshold, and the preset confidence upper limit, the independent confidence level of the water depth information source during the water entry stage is calculated. Based on the absolute value of the water depth change rate, the mean water depth, the weighting coefficient of the water depth change rate during the floating phase, the weighting coefficient of the water depth during the floating phase, the allowable rate of water depth fluctuation during the floating phase, the installation height of the vehicle's water depth sensor above the ground, and the preset confidence limit value, the independent confidence level of the water depth information source during the floating phase is calculated. Based on the water depth change rate, the mean water depth, the weighting coefficient of the water depth change rate during the water discharge stage, the water depth weighting coefficient during the water discharge stage, the water depth threshold for vehicle water discharge, the water depth descent rate threshold, and the preset confidence upper limit, the independent confidence level of the water depth information source during the water discharge stage is calculated.

10. The method according to any one of claims 1-3, characterized in that, The step of extracting features from the at least two information sources and generating an independent confidence level for each information source at each floating stage includes: If the suspension information source is included among the at least two information sources, then the vehicle front axle suspension holding force, the rate of change of the front axle suspension holding force, the vehicle rear axle suspension holding force, and the rate of change of the rear axle suspension holding force are extracted from the suspension information source. Based on the vehicle's front axle suspension holding force, the rate of change of the front axle suspension holding force, the vehicle's rear axle suspension holding force, and the rate of change of the rear axle suspension holding force, the independent confidence level of the suspension information source corresponding to each floating stage is determined.

11. The method according to claim 10, characterized in that, The determination of the independent confidence level of the suspension information source at each floating stage based on the vehicle's front axle suspension holding force, the rate of change of the front axle suspension holding force, the vehicle's rear axle suspension holding force, and the rate of change of the rear axle suspension holding force includes: Based on the vehicle's front axle suspension holding force, the rate of change of the front axle suspension holding force, the threshold of the rate of decrease of the front axle suspension holding force, the weighting coefficient of the rate of change of the front axle suspension holding force during the water entry stage, the weighting coefficient of the front axle suspension holding force, the front axle suspension holding force when the vehicle is stationary on a level road surface, and the preset confidence upper limit, the independent confidence level of the suspension information source during the water entry stage is calculated. Based on the vehicle's front axle suspension holding force, the vehicle's rear axle suspension holding force, the vehicle's front axle suspension holding force when stationary on a level road surface, the vehicle's rear axle suspension holding force when stationary on a level road surface, the weighting coefficient of the front axle suspension holding force during the floating phase, the weighting coefficient of the rear axle suspension holding force during the floating phase, and the preset confidence upper limit value, calculate the independent confidence level of the suspension information source during the floating phase. Based on the front axle suspension holding force change rate, the vehicle's front axle suspension holding force, the vehicle's front axle suspension holding force when stationary on a level road surface, the front axle suspension holding force increase rate threshold, the weighting coefficient of the front axle suspension holding force change rate during the water exit phase, the weighting coefficient of the front axle suspension holding force during the water exit phase, and the preset confidence upper limit value, the independent confidence level of the suspension information source during the water exit phase is calculated.

12. The method according to any one of claims 1-3, characterized in that, The step of extracting features from the at least two information sources and generating an independent confidence level for each information source at each floating stage includes: If the forward image information source is included among the at least two information sources, then the forward image of the vehicle is extracted from the forward image information source; Acquire the water surface area and road surface area in the image in front of the vehicle; Based on the water surface area and the road surface area, calculate the proportion of the water-involved area and the rate of change of the proportion of the water-involved area; Based on the proportion of the water-touched area and the rate of change of the proportion of the water-touched area, the independent confidence level of the forward image information source is determined for each floating stage.

13. The method according to claim 12, characterized in that, determining the independent confidence level of the foreground image information source corresponding to each floating stage based on the proportion of the wading area and the rate of change of the proportion of the wading area includes: Based on the rate of change of the proportion of the wading area, the proportion of the wading area, the weight of the rate of change of the proportion of the wading area in the front image of the vehicle during the water entry stage, the threshold for the increase of the rate of change of the proportion of the wading area, the weight of the proportion of the wading area, the saturation threshold of the proportion of the wading area and the preset confidence upper limit, the independent confidence of the front image information source during the water entry stage is calculated. Based on the proportion of the wading area, the weight of the proportion of the wading area in the vehicle's front image during the floating phase, the high proportion threshold of the wading area, and the preset confidence upper limit, the independent confidence level of the front image information source during the floating phase is calculated. Based on the change rate of the proportion of the wading area, the proportion of the wading area, the weight of the change rate of the proportion of the wading area in the vehicle's front image during the water exit stage, the threshold for the decrease in the change rate of the proportion of the wading area, the weight of the proportion of the wading area, the threshold for the low proportion of the wading area, and the preset confidence upper limit, the independent confidence level of the front image information source during the water exit stage is calculated.

14. The method according to any one of claims 1-13, characterized in that, The method further includes: The sensor signals collected by each sensor of the vehicle are acquired, including water depth signal, vehicle suspension holding force signal, vehicle front and rear axle wheel speed difference signal, front and rear axle wheel speed change signal, and vehicle pitch angle signal. Based on at least one of the water depth signal, the vehicle suspension holding force signal, the vehicle front and rear axle wheel speed difference signal, and the front and rear axle wheel speed change signal, the degree of water penetration of the vehicle's front axle is determined, and the degree of water penetration is divided into front axle not submerged, front axle slightly submerged, and front axle deeply submerged. The floating attitude of the vehicle during the floating phase is identified based at least on the degree of water entry of the front axle of the vehicle and the vehicle pitch angle signal.

15. A vehicle floating stage identification device, characterized in that, include: A construction module is used to construct at least two information sources based on data collected by various sensors of the vehicle. The at least two information sources include at least two of the following: vehicle dynamics information source, kinematics information source, water depth information source, suspension information source, and forward image information source. An extraction module is used to extract features from the at least two information sources for each floating stage, and generate an independent confidence score for each information source for each floating stage. The independent confidence score of any information source for the floating stage is used to characterize the probability that the vehicle is in the floating stage as determined by the information source. The generation module is used to merge the independent confidence scores of various information sources at the same floating stage to generate a comprehensive confidence score for each floating stage. The identification module is used to identify the current floating stage of the vehicle based on the comprehensive confidence level of each floating stage, wherein the floating stage includes the water entry stage, the floating stage, or the water exit stage.

16. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-14.

17. A vehicle, characterized in that, It includes the vehicle body and the electronic device described in claim 16.

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-14.

Images