Intelligent anti-skid control method and system for new energy vehicle

Abstract

The invention provides an intelligent anti-skid control method and system for a new energy vehicle, and belongs to the technical field of vehicle control. The method comprises the following steps: determining the type of a current macroscopic road of a vehicle through a first scene recognition model based on environmental perception information; based on the environment perception information and the vehicle state information, determining a microscopic road surface state of a current driving road surface of the vehicle through a second scene recognition model; when the microscopic road surface state meets the wet and slippery triggering condition, an anti-skid activation signal is generated, a cooperative control instruction is generated according to the macroscopic road type and the microscopic road surface state, and then a control chassis execution system is controlled to conduct anti-skid cooperative control. According to the method, the macroscopic driving scene and the microcosmic road surface state are identified in a layered manner, and the differentiated chassis cooperative control strategy is generated based on the combination of the macroscopic driving scene and the microcosmic road surface state, so that the driving safety and scene adaptability of the vehicle on the wet and slippery road surface are remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of vehicle control technology, and in particular relates to an intelligent anti-skid control method and system for new energy vehicles. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Currently, intelligent driving assistance systems in new energy vehicles have a certain degree of environmental perception capabilities. Some vehicles can even identify typical terrains such as snow, mud, and grass using sensors such as cameras, and send the identification results to the vehicle controller to achieve automatic switching of driving modes. This terrain-based mode switching function improves the vehicle's adaptability to different road surfaces to a certain extent.

[0004] However, existing technologies have the following defects and shortcomings: First, existing solutions have a relatively limited identification dimension, only reaching a coarse-grained level of terrain classification. For example, once identified as "snow" or "mud," the corresponding preset mode is invoked. This identification method cannot quantitatively assess the real-time adhesion status of the road surface, such as obtaining the specific adhesion coefficient or water depth of the current road surface, resulting in a lack of refined basis for control strategies.

[0005] Secondly, the control logic of the existing solution is relatively rigid, only realizing the switching of driving modes, without deep collaborative control with the vehicle's drive-by-wire chassis execution system. When the vehicle is on a slippery road surface, it is difficult to adjust the dynamic response parameters of the braking system, drive system and vehicle stability system in a targeted manner by simply switching modes. It is unable to implement real-time and precise anti-skid intervention according to changes in actual road conditions, and there is still a significant safety risk in complex and ever-changing slippery scenarios. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, the present invention provides an intelligent anti-skid control method and system for new energy vehicles. By hierarchically identifying macro driving scenarios and micro road conditions, and generating differentiated chassis collaborative control strategies based on the combination of the two, the driving safety and scenario adaptability of vehicles on slippery roads are significantly improved.

[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: The first aspect of this invention provides an intelligent anti-skid control method for new energy vehicles.

[0008] A smart anti-skid control method for new energy vehicles includes: Acquire environmental perception information and vehicle status information of the vehicle; Based on the environmental perception information, the macroscopic road type where the vehicle is currently located is determined by a preset first scene recognition model; the macroscopic road type includes highways, urban roads, rural roads, and non-public roads. Based on the environmental perception information and the vehicle status information, the microscopic road surface condition of the current driving surface of the vehicle is determined by a preset second scene recognition model; the microscopic road surface condition includes the estimated road surface adhesion coefficient and the thickness of the water accumulation layer; When the microscopic road surface condition meets the preset wet slip triggering condition, an anti-slip activation signal is generated; In response to the anti-skid activation signal, a coordinated control command is generated based on the macroscopic road type and the microscopic road surface condition; Based on the aforementioned coordinated control command, the vehicle's drive-by-wire chassis actuator system is controlled to perform anti-skid coordinated control.

[0009] Furthermore, the first scene recognition model also incorporates navigation map information for identifying macroscopic road types; the navigation map information includes road grade, speed limit information, and road type identifier.

[0010] Furthermore, the determination of the microscopic road surface state by the second scene recognition model includes: The system integrates and analyzes image data from a visual sensor and radar echo data from a millimeter-wave radar. It identifies reflective features, water splashes, and snow cover characteristics on the road surface using the image data. Based on the attenuation and multipath reflection characteristics of the radar echo data, it estimates the presence of water or ice on the road surface. By combining the analysis results of the image data and radar echo data, it outputs the current road surface adhesion coefficient range and water thickness level as the microscopic road surface condition.

[0011] Furthermore, based on the macroscopic road type and the microscopic road surface condition, cooperative control instructions are generated, including: Based on the microscopic road surface conditions, a basic anti-skid control strategy is determined; the basic anti-skid control strategy includes pre-filling brake wheel cylinders, adjusting the intervention threshold of the drive anti-skid system TCS, and the control parameters of the vehicle stability system ESC. Based on the aforementioned macroscopic road type, the parameters and enabling conditions of the basic anti-skid control strategy are modified to generate the cooperative control command that matches the current driving scenario.

[0012] Furthermore, the basic anti-slip control strategy is modified in terms of parameters and enable conditions, including: If the macroscopic road type is a highway, the intervention sensitivity of the vehicle stability control (ESC) system is increased, and the maximum dynamic speed limit of the vehicle is adjusted to the first speed limit value at the same time. If the macroscopic road type is an urban road, the intervention of the traction control system (TCS) is increased, and the maximum dynamic speed limit of the vehicle is adjusted to a second speed limit; the second speed limit is lower than the first speed limit. If the macroscopic road type is a rural road, the intervention of the traction control system (TCS) will be increased again, and the maximum dynamic speed limit of the vehicle will be adjusted to a third speed limit; the third speed limit is lower than the second speed limit. If the macro road type is a non-public road, the vehicle's four-wheel drive system will be forcibly switched to locked four-wheel drive mode, and the vehicle's maximum dynamic speed limit will be adjusted to the fourth speed limit, which is lower than the third speed limit.

[0013] Furthermore, after generating the anti-slip activation signal, a wet road surface warning message and an anti-slip function activation warning message are output through the human-machine interface.

[0014] Furthermore, when the anti-skid activation signal is not generated, the status of surrounding vehicles is continuously monitored; if it is detected that a surrounding vehicle loses its body posture or triggers its body stability system within a preset distance, the vehicle's position is combined with the positional relationship between the vehicle and the surrounding vehicles to predict that the vehicle is about to enter a slippery road surface area, and the vehicle's braking system and suspension system are adjusted to the ready state in advance.

[0015] A second aspect of the present invention provides an intelligent anti-skid control system for new energy vehicles.

[0016] An intelligent anti-skid control system for new energy vehicles includes: The information acquisition module is used to acquire environmental perception information and vehicle status information of the vehicle; The scene recognition module includes: The first identification unit is used to determine the macroscopic road type where the vehicle is currently located based on the environmental perception information and a preset first scene recognition model; the macroscopic road type includes highways, urban roads, rural roads, and non-public roads. The second identification unit is used to determine the microscopic road surface condition of the current driving surface of the vehicle based on the environmental perception information and the vehicle status information, through a preset second scene identification model; the microscopic road surface condition includes the estimated value of the road surface adhesion coefficient and the thickness of the water accumulation layer; The decision module is used to generate an anti-skid activation signal when the micro-road surface condition meets the preset wet-slip triggering condition; and in response to the anti-skid activation signal, generate a cooperative control command based on the macro-road type and the micro-road surface condition. The execution control module is used to control the vehicle's drive-by-wire chassis execution system to perform anti-skid collaborative control based on the collaborative control commands.

[0017] Furthermore, the second identification unit is specifically used for: The system integrates and analyzes image data from a visual sensor and radar echo data from a millimeter-wave radar. It identifies reflective features, water splashes, and snow cover characteristics on the road surface using the image data. Based on the attenuation and multipath reflection characteristics of the radar echo data, it estimates the presence of water or ice on the road surface. By combining the analysis results of the image data and radar echo data, it outputs the current road surface adhesion coefficient range and water thickness level as the microscopic road surface condition.

[0018] Furthermore, the decision-making module includes: The strategy generation submodule is used to determine a basic anti-skid control strategy based on the micro road surface conditions. The basic anti-skid control strategy includes pre-filling brake wheel cylinders, adjusting the intervention threshold of the drive anti-skid system TCS, and adjusting the control parameters of the vehicle stability system ESC. The strategy correction submodule is used to correct the parameters and enable conditions of the basic anti-skid control strategy based on the macro road type, and generate the cooperative control command that matches the current driving scenario.

[0019] The above one or more technical solutions have the following beneficial effects: This invention establishes a first scene recognition model and a second scene recognition model. The first model determines macroscopic road types such as highways, urban roads, rural roads, and non-public roads, providing scene constraints for subsequent control strategies. The second model, by fusing image data from visual sensors and radar echo data from millimeter-wave radar, quantifies and outputs microscopic road surface conditions such as the estimated road surface adhesion coefficient and the thickness of the water layer. This dual-layer recognition mechanism of "macroscopic scene qualitative + microscopic road surface quantitative" overcomes the limitations of existing technologies that rely solely on coarse-grained judgments based on terrain categories, achieving accurate perception of slippery road surface adhesion conditions and providing reliable data support for the refined adjustment of subsequent control strategies.

[0020] This invention solves the problem of rigid control logic and inability to achieve deep collaborative anti-skid intervention in existing technologies by generating collaborative control commands and driving the drive-by-wire chassis execution system. When the microscopic road surface condition meets the wet and slippery triggering conditions, this invention does not simply switch driving modes, but dynamically generates collaborative control commands by comprehensively considering the macroscopic road type and microscopic road surface condition. These commands can simultaneously act on multiple execution systems such as brake-by-wire, drive-by-wire, and vehicle stability control. For example, they can adjust the TCS intervention threshold, ESC intervention sensitivity, four-wheel drive mode, and dynamic speed limit value according to different road scenarios, achieving real-time linkage and collaborative control between multiple systems. Compared to the single control method of existing technologies that rely solely on mode switching, this invention can implement precise and adaptive anti-skid intervention based on actual road condition changes, significantly improving vehicle driving safety in complex wet and slippery scenarios.

[0021] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0022] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0023] Figure 1 This is a flowchart of an intelligent anti-skid control method for new energy vehicles according to Embodiment 1 of the present invention. Detailed Implementation

[0024] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0025] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0026] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0027] Example 1 This embodiment discloses an intelligent anti-skid control method for new energy vehicles.

[0028] like Figure 1 As shown, an intelligent anti-skid control method for new energy vehicles includes: Step S1: Obtain the vehicle's environmental perception information and vehicle status information; Step S2: Based on the environmental perception information, determine the current macroscopic road type of the vehicle using a preset first scene recognition model; the macroscopic road type includes highways, urban roads, rural roads, and non-public roads. Step S3: Based on the environmental perception information and the vehicle status information, determine the microscopic road surface state of the current driving surface of the vehicle through a preset second scene recognition model; the microscopic road surface state includes the estimated road surface adhesion coefficient and the thickness of the water accumulation layer; Step S4: When the microscopic road surface condition meets the preset wet slip triggering condition, an anti-slip activation signal is generated; Step S5: In response to the anti-skid activation signal, generate a cooperative control command based on the macroscopic road type and the microscopic road surface condition; Step S6: Based on the cooperative control command, control the vehicle's drive-by-wire chassis execution system to perform anti-skid cooperative control.

[0029] Based on the above process, this invention significantly improves vehicle driving safety and scenario adaptability on slippery roads by hierarchically identifying macroscopic driving scenarios and microscopic road surface conditions, and generating differentiated chassis cooperative control strategies based on the combination of the two. To facilitate understanding of the technical solution of this invention, the specific implementation methods of this invention will be further explained and described below.

[0030] In step S1, the vehicle's environmental perception information and vehicle status information are acquired.

[0031] Environmental perception information includes data on the vehicle's surrounding environment collected by onboard sensors. These onboard sensors include at least a forward-facing binocular camera, a surround-view camera, a rear-view camera, millimeter-wave radar, and optionally, lidar.

[0032] In practice, the forward-facing binocular camera is used to acquire image information in front of the vehicle, including lane lines, traffic signs, traffic lights, curbs, etc.; the surround-view camera and the rear-view camera are used to acquire image information on the sides and rear of the vehicle; and the millimeter-wave radar is used to acquire information such as the distance, relative speed, and radar echo intensity of obstacles in front and to the sides.

[0033] Vehicle status information includes, but is not limited to: current vehicle speed, steering wheel angle, yaw rate, longitudinal acceleration, wheel speed of each wheel, brake pedal opening, accelerator pedal opening, current driving mode, four-wheel drive status, etc. This information can be read in real time from the corresponding controllers (such as ESP, EMS, TCU, etc.) via the vehicle CAN bus or Ethernet.

[0034] In step S2, based on environmental perception information, the macroscopic road type where the vehicle is currently located is determined by a preset first scene recognition model that combines navigation map information; wherein, the macroscopic road type includes highways, urban roads, rural roads and non-public roads; the navigation map information includes road level, speed limit information and road type identification.

[0035] The first scene recognition model is deployed in the Intelligent Driving Domain Controller (ADCC). This model first extracts features from environmental perception information, such as identifying lane line types (solid lines, dashed lines), the existence of emergency lanes, the presence of traffic lights and stop lines, the speed of surrounding vehicles, and road speed limit signs.

[0036] Meanwhile, the first scene recognition model acquires high-precision map information from the in-vehicle navigation map, including the road classification of the current road segment (such as highway, national highway, provincial highway, rural road, etc.), legal speed limit, and road type identification (such as "urban expressway", "parking lot entrance and exit", etc.). The first scene recognition model integrates visual perception features with map information and determines the current macroscopic road type through a preset decision tree or classifier logic.

[0037] In this embodiment, the judgment rule is as follows: If the lane markings are clear, there is an emergency lane, there are no traffic lights or stop lines, and the speed limit is ≥80km / h, then it is considered a highway. If there are traffic lights, stop lines, and intersections, and the speed limit is between 30 and 80 km / h, it is considered an urban road. If the road is narrow, lacks clear lane markings, is surrounded by farmland or villages, and has a low speed limit (e.g., ≤40km / h), it is classified as a rural road. If a parking lot gate, continuously marked parking space, epoxy flooring, speed limit ≤20km / h, and no traffic lights or stop lines are detected, it is determined to be a non-public road (such as a parking lot or internal road of the park).

[0038] When map information and perceived features are inconsistent, the model prioritizes perceived features and records confidence levels for use in determining the reliability of subsequent strategies.

[0039] Furthermore, the first scene recognition model is a scene classifier based on multi-source information fusion. Its main components include: an input interface module, a feature extraction and alignment module, and a decision classifier module.

[0040] 1) Input interface module, responsible for receiving two types of input data.

[0041] Environmental perception information: Structured data output from visual sensors such as forward-looking cameras and surround-view cameras after processing by front-end perception algorithms, including: lane line type (solid line, dashed line, double yellow line, etc.), number of lane lines, presence of emergency lane, traffic light detection results (presence and color), stop line detection results, intersection detection results, traffic sign recognition results (speed limit sign), surrounding vehicle flow speed (estimated through target tracking), road boundaries (curb, guardrail), etc.

[0042] Navigation map information: High-precision map data from the in-vehicle navigation system, obtained in real time through the in-vehicle network, including: road level (highway, national highway, provincial highway, county road, township road, internal roads of the park, etc.), legal speed limit, road type markings (such as "urban expressway", "parking lot entrance and exit", "school area", etc.), number of lanes, whether the road is toll road, etc.

[0043] 2) Feature extraction and alignment module, used to perform spatiotemporal alignment of environmental perception features and map information.

[0044] Because perceived information may be delayed or temporarily lost, this module uses Kalman filtering to smooth the perceived features and matches the map information with the perceived information based on the vehicle's current GPS location. For example, when the vehicle travels to a section of road marked as a "highway" on the map, the module will focus on verifying whether there is an emergency lane, traffic lights, etc., in the perceived features to confirm the consistency between the map information and the actual situation.

[0045] 3) Decision classifier module, responsible for the core classification logic.

[0046] It is implemented using a pre-defined rule-based decision tree. The decision tree's judgment logic is structured based on the rules described in the initial draft, as follows: Root node: Initial screening based on map road classification. For example, if the map shows the road classification as "expressway" or "urban expressway", it will first enter the expressway road judgment branch; if it is "urban main road / secondary arterial road", it will enter the urban road judgment branch; if it is "rural road, county road", it will enter the rural road judgment branch; if there is no map information or the map shows "parking lot, park", it will enter the non-public road judgment branch.

[0047] First-level branch (perceptual feature verification): a) Highway verification: Check the perception features for the presence of an emergency lane, the clarity and continuity of lane markings, the absence of traffic lights and stop lines within a certain time and distance, and whether the surrounding vehicle speed is ≥80km / h. If the above conditions are met, the road is ultimately determined to be a highway; if not, it is downgraded to the next level of judgment (e.g., re-judged according to the speed limit).

[0048] b) Urban Road Verification: Check for the presence of traffic lights, stop lines, intersections, and clear lane markings; check the surrounding vehicle flow speed and speed limit between 30 and 80 km / h. If these conditions are met, the road is classified as an urban road; if not, downgrade it based on the map or re-evaluate it according to its features.

[0049] c) Verification of rural roads: Check whether the road width is too narrow (estimated by the distance between lane lines or the distance to the curb), whether there are surrounding environmental features such as farmland and villages (identified by semantic segmentation), whether the speed limit is ≤40km / h, and whether there are frequent occurrences of non-motorized vehicles and pedestrians. If these conditions are met, the road is determined to be a rural road.

[0050] d) Verification of non-public roads: Check for the presence of parking gates, continuously marked parking spaces, epoxy flooring characteristics (identified by road surface material), speed limit ≤20km / h, no traffic lights / stop lines, and speed bumps. If these conditions are met, the road is determined to be a non-public road (such as a parking lot or industrial park).

[0051] Second-level branch (confidence fusion and conflict handling): When there is a conflict between the perceived features and the map information (for example, the map shows a highway but the sensor does not detect the emergency lane), the module makes a judgment based on the perceived features, while reducing the output confidence and recording the state for conservative adjustments to the subsequent control strategy (for example, reducing the speed limit to ensure safety).

[0052] 4) Output interface module, used to output the current macro road type.

[0053] This module outputs the current macroscopic road type (enumerated values: highway / urban road / rural road / non-public road), and can optionally output the confidence level (0~1) for use by the subsequent strategy correction module.

[0054] In step S3, based on environmental perception information and vehicle status information, the microscopic road surface state of the current driving surface of the vehicle is determined by a preset second scene recognition model; the microscopic road surface state includes the estimated value of road surface adhesion coefficient and the thickness of water accumulation layer.

[0055] The second scene recognition model also runs in the domain controller, and its core is a multi-sensor fusion algorithm. The specific method is as follows: Image data from a visual sensor and radar echo data from a millimeter-wave radar are fused and analyzed. The visual sensor (such as a front-facing camera) acquires images of the road surface area, and a deep learning model identifies reflective features (such as specular reflection), water splash features (water mist or spray from a vehicle ahead), and snow cover features (white coverage, loss of texture). When the electromagnetic waves emitted by the millimeter-wave radar encounter water or ice, they undergo attenuation and multipath reflection, resulting in weakened echo intensity and the appearance of false targets. The second scene recognition model extracts features such as the attenuation rate and multipath reflection index of the radar echo. Combining the image recognition results with the radar features, the model outputs the current road surface adhesion coefficient range (e.g., high adhesion ≥ 0.7, medium adhesion 0.4~0.7, low adhesion ≤ 0.4) and water thickness level (e.g., no water, thin water < 2mm, medium water 2~5mm, thick water > 5mm) through a pre-defined mapping relationship (e.g., lookup table or neural network). These quantified values ​​constitute the microscopic road surface condition.

[0056] Furthermore, the second scene recognition model is a road surface condition quantification assessment model based on multi-sensor fusion. Its main components include the following modules: 1) Input interface module, used to receive input data.

[0057] Visual sensor data: raw images from a forward-looking camera (preferably a binocular camera) or feature maps processed by the front end, with a focus on the road surface area (the area of ​​the vehicle's trajectory within a certain distance in front of it).

[0058] Millimeter-wave radar data: Point cloud data from forward-looking millimeter-wave radar, including the distance, velocity, azimuth, and radar cross-section (RCS) of each target point, as well as attenuation features extracted from the raw radar echo signal.

[0059] Vehicle status information: current vehicle speed, steering wheel angle, braking, acceleration status, etc., used to assist in judgment (e.g., to eliminate the interference of the vehicle's operation on perception).

[0060] 2) Visual feature extraction module, used to process and extract features from road surface area images.

[0061] This module uses lightweight deep learning models (such as semantic segmentation networks or classification networks) to process road surface area images and extract the following features: Reflective characteristics: Identify the proportion and intensity of specular reflective areas in road surfaces. Wet and slippery road surfaces (especially those with standing water) will produce obvious specular reflections, which will appear as highlight areas in the image.

[0062] Water splash feature: Identify water splashes or mist-like water vapor caused by vehicles moving ahead. This feature can be detected by dynamic texture changes in the area surrounding the moving target.

[0063] Snow cover characteristics: Identify white covered areas on the road surface, the degree of disappearance of texture, wheel tracks, etc.

[0064] The visual feature extraction module outputs confidence scores (0~1) for the three types of features mentioned above.

[0065] 3) Radar feature extraction module, used to process millimeter-wave radar data and extract features.

[0066] Echo Attenuation Characteristics: Under normal circumstances, the intensity of radar echoes returning from the road surface has a certain statistical distribution. When there is water or ice on the road surface, electromagnetic waves are refracted or absorbed, resulting in an overall attenuation of the echo intensity. This module statistically analyzes the average echo intensity of the road surface area (after removing moving targets) and compares it with the baseline value of a dry road surface to obtain the attenuation coefficient.

[0067] Multipath reflection characteristics: Waterlogged roads create specular reflections, causing radar waves to exhibit a multipath effect after reflection, resulting in false targets (ghosting) at certain distances. This module detects the number and distribution stability of false targets as auxiliary characteristics of the presence of water.

[0068] The radar feature extraction module outputs the attenuation coefficient (0~1) and the multipath index (0~1).

[0069] 4) Fusion and quantization module, used for feature-level fusion.

[0070] This module performs feature-level fusion of visual and radar features. Specifically, it can employ one of the following two methods: Method 1 (Table Lookup Method): A mapping table is pre-established through experimental calibration. The input is a combination of visual reflectivity features, water splash features, attenuation coefficient, and multipath index. The output is the adhesion coefficient range and water thickness level. For example, if the reflectivity feature is >0.8 and the attenuation coefficient is >0.3, then the water thickness is judged to be medium (2~5mm) and the adhesion coefficient is low (≤0.4).

[0071] Method 2 (Lightweight Neural Network): The above four feature values ​​are used as inputs, passed through a 2-layer fully connected network, and output the estimated adhesion coefficient (continuous value 0~1) and the estimated water accumulation thickness (continuous value mm), which are then converted into levels through thresholding.

[0072] Based on this, as an optional implementation, the fusion and quantification module ultimately outputs the estimated value of the road surface adhesion coefficient and the thickness of the water accumulation layer. The estimated value of the road surface adhesion coefficient can be expressed as a range (e.g., ≤0.4, 0.4~0.7, ≥0.7) or a continuous value (0~1); the thickness of the water accumulation layer can be expressed as a grade (none, thin, medium, thick) or a continuous value (mm).

[0073] 5) Slippery trigger judgment module, used for comparing slippery trigger conditions.

[0074] The module receives the quantization results and compares them with preset slippery trigger conditions. As an optional embodiment, the preset conditions can be set as follows: Condition A: Estimated adhesion coefficient ≤ 0.4 (low adhesion); Condition B: Water accumulation thickness level ≥ medium (≥2mm).

[0075] When either condition A or condition B is met, the trigger module generates an anti-slip activation signal (Boolean value True); otherwise, it is False.

[0076] 6) Output interface module, used to output microscopic road surface conditions (adhesion coefficient range + water accumulation thickness level) and anti-skid activation signal for use by subsequent decision-making modules.

[0077] In step S4, when the microscopic road surface condition meets the preset wet slip triggering condition, an anti-skid activation signal is generated.

[0078] The preset wet-slip trigger condition is as follows: if the road surface adhesion coefficient is low (≤0.4) or the water accumulation thickness is medium or higher (≥2mm), the current road surface is considered wet-slip, thus meeting the trigger condition. Once the above condition is met, the domain controller's internal logic immediately generates a Boolean anti-slip activation signal and broadcasts it on the controller's internal bus for use in subsequent steps. Simultaneously, this signal can be recorded for fault diagnosis.

[0079] In step S5, in response to the anti-skid activation signal, a coordinated control command is generated based on the macroscopic road type and microscopic road surface condition. This can be implemented using the following methods: First, based on the aforementioned microscopic road surface conditions, a basic anti-skid control strategy is determined. This strategy includes pre-filling brake wheel cylinders, adjusting the intervention threshold of the traction control system (TCS), and the control parameters of the electronic stability control system (ESC). Specifically: 1) Pre-filled brake wheel cylinders: The domain controller sends instructions to the braking system (such as ONEbox) to pre-build a certain braking pressure (e.g., 5~10 bar) to eliminate brake clearance and shorten the response time during emergency braking.

[0080] 2) Adjust the TCS intervention threshold: Reduce the slip rate threshold of the TCS from the default value (e.g., 20%) to a lower value (e.g., 10%), so that the TCS intervenes earlier to prevent excessive slippage of the drive wheels.

[0081] 3) Adjust ESC control parameters: Appropriately increase the yaw rate control gain of ESC so that ESC can intervene with single-wheel braking more quickly when the vehicle shows signs of instability.

[0082] Secondly, based on the aforementioned macroscopic road type, the basic anti-skid control strategy is modified in terms of parameters and enabling conditions to generate the cooperative control command that matches the current driving scenario. The specific modification method is as follows: If the macroscopic road type is a highway, the intervention sensitivity of ESC should be further increased (for example, by increasing the yaw rate control gain by 30%), while the maximum dynamic speed limit of the vehicle should be adjusted to the first speed limit (for example, 80 km / h). This is because highways have high vehicle speeds and extremely high stability requirements on slippery surfaces, so ESC needs to intervene more actively, while limiting the maximum speed to prevent loss of control.

[0083] If the macro-road type is urban road, then the intervention of TCS should be increased (e.g., the TCS slip rate threshold should be further reduced to 8%), while the maximum dynamic speed limit for vehicles should be adjusted to the second speed limit (e.g., 50 km / h), which is lower than the first speed limit. Urban roads have relatively low vehicle speeds but dense pedestrian and vehicle traffic, so it is necessary to prioritize ensuring that the drive wheels do not slip, and the speed limit should be more stringent.

[0084] If the macroscopic road type is rural road, then the intervention intensity of TCS should be increased again (for example, by reducing the TCS slip rate threshold to 5%), while adjusting the maximum dynamic speed limit of the vehicle to a third speed limit (for example, 30 km / h), which is lower than the second speed limit. Rural roads have worse surface conditions, and the coefficient of friction may be extremely low, requiring TCS to be extremely sensitive, and the speed limit is even lower.

[0085] If the main road type is a non-public road (such as a parking lot), the vehicle's four-wheel drive system will be forcibly switched to locked four-wheel drive mode (if the vehicle is equipped with on-demand four-wheel drive or electric four-wheel drive) to ensure that all four wheels have driving force. At the same time, the vehicle's maximum dynamic speed limit will be adjusted to the fourth speed limit (e.g., 15 km / h), which is lower than the third speed limit. Non-public roads are usually narrow and slippery (such as epoxy flooring), and locked four-wheel drive can provide maximum traction, ensuring safety at extremely low speeds.

[0086] The aforementioned correction parameters (such as speed limit values ​​and threshold adjustment amounts) can be pre-stored in the parameter table of the domain controller and directly called according to the road type.

[0087] In step S6, based on the cooperative control command, the vehicle's drive-by-wire chassis actuator system is controlled to perform anti-skid cooperative control.

[0088] The domain controller sends coordinated control commands via the vehicle network to the ONEbox braking system, the steer-by-wire system (motor controller), the EPS steering system, and the active suspension system (if equipped). The braking system establishes pressure based on pre-fill commands and performs stability control by adjusting parameters according to ESC. The drive system limits motor output torque based on TCS parameters and speed limits, and forcibly distributes torque between the front and rear axles in four-wheel drive lock-up mode. The steering system can appropriately adjust steering assist characteristics to provide clearer road feel. The active suspension system can adjust shock absorber damping to improve tire contact. All execution systems work together to achieve vehicle stability control on slippery surfaces.

[0089] Furthermore, after generating the anti-slip activation signal, a wet road surface warning message and an anti-slip function activation warning message are output through the human-machine interface.

[0090] In practice, the human-machine interface includes the instrument panel display and the central control screen. When the anti-slip activation signal is active, the instrument panel displays a "slippery road surface" icon (such as a water droplet shape or snowflake pattern) and the text "Intelligent anti-slip activated." Simultaneously, the central control screen displays a pop-up notification informing the driver that the system has entered slippery mode and the maximum speed is limited. These notifications help the driver understand the vehicle's status and avoid accidental operation.

[0091] Furthermore, when the anti-skid activation signal is not generated, the status of surrounding vehicles is continuously monitored; if it is detected that a surrounding vehicle loses its body posture or triggers its body stability system within a preset distance, the vehicle's position is combined with the positional relationship between the vehicle and the surrounding vehicles to predict that the vehicle is about to enter a slippery road surface area, and the vehicle's braking system and suspension system are adjusted to the ready state in advance.

[0092] In practice, the domain controller continuously tracks the movement of vehicles ahead using a forward-facing camera and millimeter-wave radar. If a vehicle ahead (e.g., within 50-100 meters) is detected to suddenly skid, fishtail, or brake suddenly (which can be determined by analyzing its lateral displacement speed, yaw angle changes, or brake light illumination), and this instability is suspected to be caused by a slippery road surface, the vehicle anticipates entering the same slippery area. At this point, the domain controller sends a pre-instruction to the braking system, pre-filling the brake wheel cylinders to a low-pressure ready state, and also sends a command to the active suspension system to adjust the shock absorber damping to a stiffer mode to enhance support, preparing for potential emergency maneuvers. This predictive function provides additional safety when the vehicle's sensors have not yet directly detected a slippery road surface.

[0093] Example 2 This embodiment discloses an intelligent anti-skid control system for new energy vehicles.

[0094] An intelligent anti-skid control system for new energy vehicles includes: The information acquisition module is used to acquire environmental perception information and vehicle status information of the vehicle; The scene recognition module includes: The first identification unit is used to determine the macroscopic road type where the vehicle is currently located based on the environmental perception information and a preset first scene recognition model; the macroscopic road type includes highways, urban roads, rural roads, and non-public roads. The second identification unit is used to determine the microscopic road surface condition of the current driving surface of the vehicle based on the environmental perception information and the vehicle status information, through a preset second scene identification model; the microscopic road surface condition includes the estimated value of the road surface adhesion coefficient and the thickness of the water accumulation layer; The decision module is used to generate an anti-skid activation signal when the micro-road surface condition meets the preset wet-slip triggering condition; and in response to the anti-skid activation signal, generate a cooperative control command based on the macro-road type and the micro-road surface condition. The execution control module is used to control the vehicle's drive-by-wire chassis execution system to perform anti-skid collaborative control based on the collaborative control commands.

[0095] Furthermore, the second identification unit is specifically used for: The system integrates and analyzes image data from a visual sensor and radar echo data from a millimeter-wave radar. It identifies reflective features, water splashes, and snow cover characteristics on the road surface using the image data. Based on the attenuation and multipath reflection characteristics of the radar echo data, it estimates the presence of water or ice on the road surface. By combining the analysis results of the image data and radar echo data, it outputs the current road surface adhesion coefficient range and water thickness level as the microscopic road surface condition.

[0096] Furthermore, the decision-making module includes: The strategy generation submodule is used to determine a basic anti-skid control strategy based on the micro road surface conditions. The basic anti-skid control strategy includes pre-filling brake wheel cylinders, adjusting the intervention threshold of the drive anti-skid system TCS, and adjusting the control parameters of the vehicle stability system ESC. The strategy correction submodule is used to correct the parameters and enable conditions of the basic anti-skid control strategy based on the macro road type, and generate the cooperative control command that matches the current driving scenario.

[0097] Furthermore, the strategy correction submodule is specifically used for: If the macroscopic road type is a highway, the intervention sensitivity of the vehicle stability control (ESC) system is increased, and the maximum dynamic speed limit of the vehicle is adjusted to the first speed limit value at the same time. If the macroscopic road type is an urban road, the intervention of the traction control system (TCS) is increased, and the maximum dynamic speed limit of the vehicle is adjusted to a second speed limit; the second speed limit is lower than the first speed limit. If the macroscopic road type is a rural road, the intervention of the traction control system (TCS) will be increased again, and the maximum dynamic speed limit of the vehicle will be adjusted to a third speed limit; the third speed limit is lower than the second speed limit. If the macro road type is a non-public road, the vehicle's four-wheel drive system will be forcibly switched to locked four-wheel drive mode, and the vehicle's maximum dynamic speed limit will be adjusted to the fourth speed limit, which is lower than the third speed limit.

[0098] Furthermore, an intelligent anti-skid control system for new energy vehicles also includes a human-machine interaction module, used to output slippery road surface warning information and / or anti-skid function activation warning information after generating the anti-skid activation signal.

[0099] Furthermore, the scene recognition module also includes: The prediction unit is used to continuously monitor the status of surrounding vehicles when the anti-skid activation signal is not generated; if it detects that a surrounding vehicle loses its body posture or triggers its body stability system within a preset distance, it predicts that the vehicle is about to enter a slippery road surface area based on the position relationship between the vehicle and the surrounding vehicles, and adjusts the vehicle's braking system and / or suspension system to a ready state in advance.

[0100] Furthermore, the steer-by-wire chassis execution system includes: a brake-by-wire system, a steering-by-wire system, a drive-by-wire system, and an active suspension system.

[0101] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0102] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. An intelligent anti-skid control method for new energy vehicles, characterized in that, include: Acquire environmental perception information and vehicle status information of the vehicle; Based on the environmental perception information, the macroscopic road type where the vehicle is currently located is determined by a preset first scene recognition model; the macroscopic road type includes highways, urban roads, rural roads, and non-public roads. Based on the environmental perception information and the vehicle status information, the microscopic road surface state of the current driving road surface of the vehicle is determined by a preset second scene recognition model; The microscopic pavement condition includes the estimated pavement adhesion coefficient and the thickness of the water accumulation layer; When the microscopic road surface condition meets the preset wet slip triggering condition, an anti-slip activation signal is generated; In response to the anti-skid activation signal, a coordinated control command is generated based on the macroscopic road type and the microscopic road surface condition; Based on the aforementioned coordinated control command, the vehicle's drive-by-wire chassis actuator system is controlled to perform anti-skid coordinated control.

2. The intelligent anti-skid control method for new energy vehicles as described in claim 1, characterized in that, The first scene recognition model combines macro-road type identification with navigation map information; the navigation map information includes road grade, speed limit information and road type identifier.

3. The intelligent anti-skid control method for new energy vehicles as described in claim 1, characterized in that, The second scene recognition model determines the microscopic road surface state, including: The system integrates and analyzes image data from a visual sensor and radar echo data from a millimeter-wave radar. It identifies reflective features, water splashes, and snow cover characteristics on the road surface using the image data. Based on the attenuation and multipath reflection characteristics of the radar echo data, it estimates the presence of water or ice on the road surface. By combining the analysis results of the image data and radar echo data, it outputs the current road surface adhesion coefficient range and water thickness level as the microscopic road surface condition.

4. The intelligent anti-skid control method for new energy vehicles as described in claim 1, characterized in that, Based on the macroscopic road type and the microscopic road surface condition, a cooperative control command is generated, including: Based on the microscopic road surface conditions, a basic anti-skid control strategy is determined; the basic anti-skid control strategy includes pre-filling brake wheel cylinders, adjusting the intervention threshold of the drive anti-skid system TCS, and the control parameters of the vehicle stability system ESC. Based on the aforementioned macroscopic road type, the parameters and enabling conditions of the basic anti-skid control strategy are modified to generate the cooperative control command that matches the current driving scenario.

5. The intelligent anti-skid control method for new energy vehicles as described in claim 4, characterized in that, The basic anti-slip control strategy is modified in terms of parameters and enable conditions, including: If the macroscopic road type is a highway, the intervention sensitivity of the vehicle stability control (ESC) system is increased, and the maximum dynamic speed limit of the vehicle is adjusted to the first speed limit value at the same time. If the macroscopic road type is an urban road, the intervention of the anti-skid system TCS is increased, and the maximum dynamic speed limit of the vehicle is adjusted to a second speed limit; the second speed limit is lower than the first speed limit. If the macroscopic road type is a rural road, the intervention of the traction control system (TCS) will be increased again, and the maximum dynamic speed limit of the vehicle will be adjusted to a third speed limit; the third speed limit is lower than the second speed limit. If the macro road type is a non-public road, the vehicle's four-wheel drive system will be forcibly switched to locked four-wheel drive mode, and the vehicle's maximum dynamic speed limit will be adjusted to the fourth speed limit, which is lower than the third speed limit.

6. The intelligent anti-skid control method for new energy vehicles as described in claim 1, characterized in that, After generating the anti-slip activation signal, a slippery road surface warning message and an anti-slip function activation warning message are output through the human-machine interface.

7. The intelligent anti-skid control method for new energy vehicles as described in claim 1, characterized in that, When the anti-skid activation signal is not generated, the status of surrounding vehicles is continuously monitored; if it is detected that a surrounding vehicle loses its body posture or triggers its body stability system within a preset distance, the vehicle's position is combined with the positional relationship between the vehicle and the surrounding vehicles to predict that the vehicle is about to enter a slippery road surface area, and the vehicle's braking system and suspension system are adjusted to the ready state in advance.

8. An intelligent anti-skid control system for new energy vehicles, characterized in that, include: The information acquisition module is used to acquire environmental perception information and vehicle status information of the vehicle; The scene recognition module includes: The first identification unit is used to determine the macroscopic road type where the vehicle is currently located based on the environmental perception information and a preset first scene recognition model; the macroscopic road type includes highways, urban roads, rural roads, and non-public roads. The second identification unit is used to determine the microscopic road surface condition of the current driving surface of the vehicle based on the environmental perception information and the vehicle status information, through a preset second scene identification model; the microscopic road surface condition includes the estimated value of the road surface adhesion coefficient and the thickness of the water accumulation layer; The decision module is used to generate an anti-skid activation signal when the micro-road surface condition meets the preset wet-slip triggering condition; and in response to the anti-skid activation signal, generate a cooperative control command based on the macro-road type and the micro-road surface condition. The execution control module is used to control the vehicle's drive-by-wire chassis execution system to perform anti-skid collaborative control based on the collaborative control commands.

9. The intelligent anti-skid control system for new energy vehicles as described in claim 8, characterized in that, The second identification unit is specifically used for: The system integrates and analyzes image data from a visual sensor and radar echo data from a millimeter-wave radar. It identifies reflective features, water splashes, and snow cover characteristics on the road surface using the image data. Based on the attenuation and multipath reflection characteristics of the radar echo data, it estimates the presence of water or ice on the road surface. By combining the analysis results of the image data and radar echo data, it outputs the current road surface adhesion coefficient range and water thickness level as the microscopic road surface condition.

10. The intelligent anti-skid control system for new energy vehicles as described in claim 8, characterized in that, The decision-making module includes: The strategy generation submodule is used to determine a basic anti-skid control strategy based on the micro road surface conditions. The basic anti-skid control strategy includes pre-filling brake wheel cylinders, adjusting the intervention threshold of the drive anti-skid system TCS, and adjusting the control parameters of the vehicle stability system ESC. The strategy correction submodule is used to correct the parameters and enable conditions of the basic anti-skid control strategy based on the macro road type, and generate the cooperative control command that matches the current driving scenario.

Images