A vehicle height limit warning method and system

Abstract

The application relates to the technical field of intelligent driving, and discloses a vehicle height-limiting early warning method and system, which comprises the following steps: acquiring three-dimensional environment data, self-state data and cooperative geometric information received through vehicle-to-vehicle communication of a vehicle; constructing a final dynamic vehicle model taking into account real-time attitudes of the vehicle, load changes and future dynamic bumping responses predicted based on a road surface profile; fusing three-dimensional perception data of the vehicle and the cooperative geometric information to generate an enhanced three-dimensional environment model. Predictive interference analysis is performed on the final dynamic vehicle model and the enhanced three-dimensional environment model on a predicted driving path, and a graded early warning is output to a driver. The application accurately depicts the future space-time relationship between the vehicle and the environment by constructing a high-precision dynamic vehicle model and an information-enhanced environment model, and breaks through the limitations of single-vehicle detection with the aid of cooperative perception, so that reliable early warning can be performed on long-distance, shielded or dynamically changing height-limiting risks.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, specifically to a method and system for warning of vehicle height restrictions. Background Technology

[0002] With the rapid development of intelligent transportation systems and autonomous driving technology, vehicle safety has become one of the core issues in technological research and development. Among these issues, accidents involving vehicles exceeding height limits and colliding with bridges, tunnels, culverts, gantries, and other height-restricted facilities are particularly frequent for large trucks, buses, and special-purpose vehicles, often resulting in severe property damage, traffic congestion, and even casualties. Therefore, developing a reliable vehicle height restriction warning system has significant practical and commercial value.

[0003] Currently, early warning technologies for the risk of vehicle height restrictions mainly include the following methods:

[0004] The first type is warnings based on fixed roadside facilities, such as physical height restriction bars or optical / infrared detection devices placed in front of height restriction points. While this method is direct, its coverage is limited, only providing warnings for a single point of imminent passage, and failing to offer drivers sufficient reaction time and decision-making space for route replanning. Furthermore, its construction and maintenance costs are high, making widespread deployment difficult.

[0005] The second type is alerts based on electronic map data. Many commercial navigation systems, especially those developed for trucks, pre-load information on known height restrictions into their map data. When a vehicle's planned route passes through these height restrictions, the system provides a warning based on the user-inputted vehicle height. However, the effectiveness of this technology heavily relies on the accuracy and timeliness of the map data. Map data may be outdated, lack sufficient accuracy, or fail to include temporary height restriction facilities (such as gantries in construction areas). More critically, this method typically uses a static, fixed vehicle height value for judgment, completely ignoring the dynamic changes in vehicle posture (such as pitch, roll, and suspension undulation) caused by factors such as load variations, uneven road surfaces, and acceleration and braking during actual driving. This can lead to serious misjudgments or omissions at critical heights.

[0006] The third type is real-time detection based on onboard sensors. Some vehicles are equipped with sensors such as ultrasonic or lidar pointing towards the top or upper front of the vehicle to detect obstacles above in real time. This method can detect obstacles not marked on the map, but its inherent limitations are also quite obvious: the detection range is very limited. It is essentially a last-ditch warning before a collision, rather than an early warning before a risk, leaving the driver with very little reaction time, often making it impossible to avoid an accident. At the same time, single-vehicle perception is limited by obstructions to the line of sight and cannot detect risk points obstructed by vehicles behind curves or in front.

[0007] In summary, existing technologies either rely on incomplete prior data, ignore the dynamic characteristics of vehicles, or have insufficient detection range and lack predictability, making it difficult to provide a comprehensive, accurate, and timely warning of height restriction risks. Therefore, how to comprehensively utilize multiple information sources to construct a dynamic model that can accurately reflect the real-time space occupancy of vehicles and make long-distance, highly reliable predictions of height restriction risks along future driving paths has become an urgent technical problem to be solved in this field. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides a vehicle height restriction warning method and system, which solves the problems of inaccurate, untimely, and unreliable height restriction warnings caused by the neglect of dynamic changes in vehicle movement and limited perception capabilities in existing technologies.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a vehicle height restriction warning method, comprising the following steps:

[0010] Acquire real-time status data of the vehicle, including real-time attitude information and real-time load information of the vehicle;

[0011] Acquire three-dimensional environmental data in front of the vehicle;

[0012] Based on the real-time status data, a dynamic vehicle model representing the three-dimensional space occupied by the vehicle is generated;

[0013] Based on the three-dimensional environmental data, a three-dimensional environmental model representing the passable space in front of the vehicle is constructed;

[0014] Interference analysis is performed on the dynamic vehicle model and the three-dimensional environment model to determine whether there is a risk of height restriction.

[0015] Preferably, before the step of performing interferometric analysis, the method further includes:

[0016] Extract the road surface contour along the path in front of the vehicle from the three-dimensional environment model;

[0017] The instantaneous vertical displacement of the vehicle caused by road bumps is predicted based on the road surface profile.

[0018] The dynamic vehicle model is then corrected based on the instantaneous vertical displacement.

[0019] Preferably, the step of performing interference analysis on the dynamic vehicle model and the three-dimensional environment model specifically includes:

[0020] The modified dynamic vehicle model is virtually simulated along the path in front of the vehicle, and it is determined whether there is spatial overlap between the model and the boundary of the three-dimensional environment model.

[0021] Preferably, the step of constructing a three-dimensional environment model representing the passable space in front of the vehicle includes:

[0022] Based on the aforementioned three-dimensional environment data, an initial three-dimensional environment model is constructed;

[0023] It receives cooperative geometric information broadcast via vehicle-to-vehicle communication and uses the cooperative geometric information to update the initial three-dimensional environment model to obtain the three-dimensional environment model.

[0024] Preferably, when the interference analysis determines that there is a height restriction risk caused by a newly detected obstacle of the vehicle, the method encodes the geometric data of the obstacle into cooperative geometric information and broadcasts it through vehicle-to-vehicle communication.

[0025] Preferably, the step of generating the dynamic vehicle model specifically includes:

[0026] For the preset vehicle base model, a rotation transformation based on the real-time attitude information and a translation transformation based on the real-time load information are applied sequentially.

[0027] Preferably, the step of constructing a three-dimensional environment model representing the passable space in front of the vehicle specifically includes:

[0028] The road surface is identified and fitted from the three-dimensional environmental data;

[0029] The non-road surface 3D environment data is identified as obstacles, and the boundaries of the 3D environment model are defined by the obstacles.

[0030] Preferably, when the interference analysis determines that there is a height restriction risk, the method executes different levels of warning prompts based on the distance between the height restriction risk and the vehicle.

[0031] Preferably, the step of constructing a three-dimensional environment model representing the passable space in front of the vehicle includes:

[0032] Based on the aforementioned three-dimensional environment data, an initial three-dimensional environment model is constructed;

[0033] It receives cooperative geometric information broadcast via vehicle-to-vehicle communication and uses the cooperative geometric information to update the initial three-dimensional environment model to obtain the three-dimensional environment model.

[0034] A vehicle height restriction warning system includes:

[0035] A processor and a memory connected to the processor, wherein the memory stores program instructions, and when the processor executes the program instructions, it implements a vehicle height restriction warning method.

[0036] This invention provides a vehicle height restriction warning method and system. It has the following beneficial effects:

[0037] 1. This invention generates a dynamic vehicle model that not only accounts for changes in the vehicle's geometric shape caused by real-time load and driving posture, but also predicts and incorporates the instantaneous vertical displacement of the vehicle caused by road bumps by analyzing the road surface profile ahead. This makes the three-dimensional spatial representation of the vehicle closer to physical reality, thus making the final interferometry analysis results more accurate than the method using static height values.

[0038] 2. This invention actively constructs a 3D environment model using an onboard 3D perception device, without relying on pre-stored map data that may have update delays or missing information. Furthermore, by receiving collaborative geometric information broadcast via vehicle-to-vehicle communication, the system can integrate obstacle data detected by other vehicles, far exceeding the vehicle's perception range, into the local model, thereby achieving rapid and collaborative perception of sudden height restriction risks such as temporary construction and accidents.

[0039] 3. This invention employs a predictive interferometry method to virtually simulate the forward path along a meticulously modeled dynamic vehicle model in both time and space. This method does not only detect hazards when the vehicle is near a danger point, but also predicts risks from a greater distance by calculating whether the vehicle model and the environment model will spatially overlap in the future. This allows the driver more time to make a decision regarding braking or evasive maneuvers. Attached Figure Description

[0040] Figure 1 This is a system structure block diagram of the present invention;

[0041] Figure 2 This is a flowchart of the height restriction warning method of the present invention. Detailed Implementation

[0042] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Please see the appendix Figure 1 -Appendix Figure 2 This invention provides a vehicle height restriction warning system, comprising:

[0044] The processing unit is a set of internally fixed program logic configured to execute a series of tightly coupled calculation and analysis steps in order to predict the risk of height restrictions.

[0045] During system operation, the core task of the processing unit begins with the reception and processing of multi-source heterogeneous data. It continuously acquires data streams with synchronization timestamps from the data acquisition unit, including high-density 3D environmental point clouds output by the 3D perception module, and real-time pitch angle θ, roll angle φ, and vehicle vertical displacement Δz output by the vehicle state perception module. load This includes the vehicle's precise location, speed, and predicted driving path output by the positioning module. Time alignment of this data is fundamental to ensuring the accuracy of subsequent analysis.

[0046] After receiving the data, a dynamic vehicle model can be generated. The purpose of this model is to generate a digital copy that accurately reflects the instantaneous three-dimensional spatial occupancy of the vehicle at the present and future moments. This process first retrieves a locally stored basic three-dimensional geometric model M of the vehicle. base The model defines the standard shape of the vehicle in an unloaded, semi-stationary state in water. The processing unit then constructs a rotation matrix R based on the real-time attitude angles θ and φ. pose , for M base The vehicle's posture changes to simulate its spatial form when climbing or tilting.

[0047] Based on the vertical displacement Δz of the vehicle body caused by the real-time load load Construct translation vector T load The model, having undergone attitude transformation, is vertically translated to account for the impact of load changes on vehicle height. After these two steps, a preliminary dynamic vehicle model reflecting quasi-static effects is generated. Parallel to the vehicle model construction, the processing unit is also responsible for building the 3D environment model. This step aims to establish an accurate model describing the safe passage space ahead based on the vehicle's real-time perception.

[0048] The processing unit applies the Random Sample Consensus (RANSAC) algorithm to process the original 3D environment point cloud, efficiently segmenting the road surface point set and fitting a high-precision road surface model S. road All non-road point clouds are classified as the potential obstacle point set O. lidar Subsequently, the processing unit, along the predicted driving path and within a preset lateral safety width, performs Boolean operations to convert O... lidar The space occupied is subtracted from an infinitely high initial free space to obtain a locally perceived virtual security channel model.

[0049] The processing unit further refines and enhances the generated model. To account for the dynamic bumps caused by uneven road surfaces during actual driving, the processing unit performs path-related corrections on the dynamic vehicle model. This begins with the already constructed road surface model S. roadIn the process, the elevation profile function h on the predicted vehicle driving path is extracted. road (s), where s is the arc length along the path. This profile function is used as the excitation input for a pre-defined vehicle suspension system dynamics model, which can be described by the following second-order differential equation:

[0050]

[0051] Where m, c, and k are the equivalent mass, equivalent damping coefficient, and equivalent spring stiffness of the vehicle suspension system, respectively.

[0052] y(t) is the value derived from the road surface elevation h. road (s) and the excitation input at the wheel where the vehicle speed is determined;

[0053] z(t) is the instantaneous vertical displacement response of the vehicle body obtained by the processing unit through numerically solving the equation.

[0054] The processing unit will process this displacement Δz, which varies with the path position s. dyn (s) is then applied again to the previously generated preliminary dynamic vehicle model using a translation transformation, ultimately resulting in a final dynamic vehicle model M capable of predicting future dynamic bumps. final (s).

[0055] To overcome the limitations of single-vehicle perception, the 3D environment model is also enhanced through collaborative methods. The processing unit continuously listens to and parses the collaborative geometric information G broadcast by surrounding vehicles via the communication unit. v2v This information includes three-dimensional geometric models of height-restricted obstacles detected by other vehicles, which may be far beyond the perception range of this vehicle. v2v The processing unit will model these external obstacles O v2v Obstacle point set O perceived by this vehicle lidar Merging them in three-dimensional space forms a more complete global obstacle set O. total This is then used to update the virtual secure passage model, ultimately resulting in an enhanced 3D environment model V. enhanced .

[0056] Once the final dynamic vehicle model and the enhanced 3D environment model are ready, the processing unit performs the core predictive interferometry analysis. This step couples the two refined models in the spatiotemporal dimensions. Specifically, the processing unit couples the vehicle model M, which dynamically changes with the path position s, with the path position s. final ( s ) A virtual simulation is performed along the predicted driving path.

[0057] Each location point s along this deduction path kEach processing unit performs a rapid spatial interference detection, and its mathematical judgment condition is: (The parameters of this formula are defined in the previous text), that is, to determine whether the volume occupied by the vehicle model at this location point overlaps with the volume occupied by the global obstacle set.

[0058] Once the interference analysis determines that a potential risk exists, the processing unit immediately assesses the urgency of the risk and assigns a warning level based on the distance between the location of the initial interference and the vehicle. It then generates a corresponding instruction and sends it to the human-machine interface unit to issue a warning to the driver using one or more combined methods such as sound, light, or touch.

[0059] Furthermore, if the obstacle causing the risk is confirmed to be newly discovered by the vehicle, meaning that its information is not present in any received cooperative geometric information, the processing unit will also be responsible for packaging and encoding the precise three-dimensional geometric data of the obstacle and instructing the communication unit to broadcast it outward, thereby realizing early warning to surrounding vehicles and improving the overall safety of regional traffic.

[0060] In this embodiment, the data acquisition unit provides comprehensive and accurate raw data input for the processing unit's analysis and decision-making.

[0061] Specifically, it includes a 3D perception module configured to acquire 3D environmental data in front of the vehicle. Preferably, this module can be a forward-facing long-range lidar (LiDAR). During vehicle movement, this module continuously emits laser beams into a forward fan-shaped area and receives reflected signals. By calculating the signal flight time, it accurately determines the 3D coordinates of a large number of points on the obstacle surface, thereby generating a high-density 3D environmental point cloud. This point cloud data is the primary basis for the processing unit to construct a 3D environmental model, particularly for identifying road surface contours and potential height-restricted obstacles.

[0062] The data acquisition unit also includes a vehicle state perception module to acquire real-time data characterizing the vehicle's physical state. This module further includes an inertial measurement unit (IMU) configured to measure and output the pitch angle θ and roll angle φ, representing the vehicle's current attitude, in real time. This angular data is transmitted to a processing unit as a direct basis for attitude rotation transformation of the vehicle's basic three-dimensional geometric model, enabling the dynamic vehicle model to accurately reflect attitude changes caused by climbing, cornering, or road inclination. The vehicle state perception module also includes a component for measuring load changes, which may be a load sensor or an interface connected to the vehicle's air suspension system controller. This component's function is to acquire in real-time the change in vehicle height caused by cargo loading / unloading or changes in passenger numbers and output it as a specific vertical displacement value Δz. loadThis data is used by the processing unit to perform a vertical translation transformation on the vehicle model to ensure that the height of the dynamic vehicle model accurately reflects the current load state. In addition, this data acquisition unit also includes a positioning module.

[0063] This module not only provides the vehicle's high-precision position in the global coordinate system but also outputs the vehicle's real-time speed and generates a high-confidence short-term predicted driving path based on navigation information. This path information serves as the baseline trajectory for the processing unit to perform virtual simulations during predictive interferometry analysis; while the real-time speed is a key parameter used by the processing unit to convert the road surface profile in the spatial domain into the excitation input in the time domain when solving the suspension system dynamics model.

[0064] To ensure the effective fusion of multi-source data, all data output by the data acquisition unit, or in the initial stage of data reception by the processing unit, is assigned a unified high-precision timestamp.

[0065] Since the operating frequencies and data generation sequences of each sensor module differ, timestamp synchronization is a necessary step to ensure that the environmental point cloud, vehicle attitude, load information, and position processed by the processing unit at any given moment correspond to the same physical instant. This is crucial for the accuracy of subsequent modeling and analysis.

[0066] The communication unit constructs independent vehicle nodes into a collaborative perception network, enabling the system to acquire environmental information beyond the detection range of the vehicle's own sensors.

[0067] The communication unit is used for receiving information and is configured to continuously listen for and receive cooperative geometric information broadcast via a vehicle-to-vehicle (V2V) communication channel. This information is a data packet carrying rich geometric details. The data packet encapsulates specific geometric data of height-restricted obstacles detected by other vehicles, such as the obstacle's precise three-dimensional coordinates, size boundaries (such as length, width, and height), and even point sets or grid data describing its contours.

[0068] Upon receiving this data packet, the communication unit performs preliminary parsing and transmits the structured geometric data to the processing unit. The processing unit then fuses this external obstacle information with the local environment model, which is constructed in real-time by the vehicle's 3D perception module, in three-dimensional space. This dynamically enhances the system's 3D environment model, incorporating height-restricted risk points that are far ahead, obstructed, or difficult for the vehicle's sensors to detect stably in adverse weather conditions. This provides a data foundation for the processing unit to perform longer-term, more reliable predictive interferometry analysis.

[0069] Another core function of this communication unit is information transmission. This broadcasting behavior is not triggered unconditionally, but is driven by the processing unit based on specific logic. During system operation, when the processing unit determines, in its predictive interference analysis, that a newly detected obstacle poses a height restriction risk, and confirms through comparison that the obstacle information is not yet included in any received cooperative geometry information, then the obstacle is considered a new risk.

[0070] In this scenario, the processing unit extracts the precise three-dimensional geometric data of the new obstacle and encodes it into a new cooperative geometric information according to a preset protocol format. The processing unit then instructs the communication unit to broadcast this encoded data packet. Upon receiving the instruction, the communication unit transmits this information to other vehicles in the surrounding area via its V2V communication interface.

[0071] This vehicle becomes an information contributing node in the collaborative sensing network. Its detection results can provide timely and high-value height restriction warning information for other vehicles passing through this section of road, thereby improving road traffic safety in a wider range.

[0072] The human-machine interaction unit transforms the abstract risk assessment results output by the processing unit into warning signals that the driver can intuitively perceive and understand.

[0073] The human-machine interface unit receives instructions from the processing unit. These instructions not only assess the presence or absence of risk, but more importantly, they include a quantitative rating of the risk's urgency. The processing unit determines the current risk level based on the distance between the vehicle and potential interference points calculated through its predictive interference analysis. The human-machine interface unit is configured to execute warning strategies of varying intensities or combinations based on this risk level. This tiered warning mechanism aims to provide appropriate alerts based on the actual proximity of the danger, avoiding excessive interference when the risk is still far away, and ensuring that a sufficiently strong warning is issued to attract the driver's attention when the danger is imminent.

[0074] This human-computer interaction unit can integrate multiple warning output channels, including visual, auditory, and tactile channels.

[0075] For visual warning channels, this can be implemented on the vehicle's central control display or head-up display system. When the processing unit determines that a low-level long-distance risk exists, the human-machine interface unit can mark the risk location on the navigation map with a specific icon or highlighted path segment. As the risk level increases, the visual warning will be strengthened accordingly; for example, the icon will turn bright red and begin to flash, while a clear warning window will pop up on the screen. This window can further display a simulated image of the obstacle, the calculated minimum safe passage height, and the real-time distance to the risk point.

[0076] The auditory warning system is implemented through the vehicle's speaker system. Corresponding to visual warnings, the intensity of the auditory warning is also linked to the risk level. Initial low-risk levels may only trigger a soft warning sound. As the vehicle continues to approach the risk point and the risk level increases, the frequency and volume of the warning sound will increase, or it may transform into a continuous, more alarming buzzing sound. At the highest risk level, the system can also broadcast preset voice messages, such as "Height restriction risk 100 meters ahead, please slow down immediately."

[0077] To provide a warning mechanism that remains effective even when the driver's visual and auditory attention is distracted, the human-machine interface unit may also include a tactile warning channel. This channel can be implemented via a vibration motor integrated into the driver's seat. High-risk commands issued by the processing unit trigger seat vibrations; the intensity or pulse pattern of the vibration is directly correlated with the urgency of the risk. This direct physical stimulus can overcome visual and auditory distractions, delivering a clear and unambiguous danger signal to the driver.

[0078] In a specific embodiment, the vehicle height restriction warning method may include the following steps:

[0079] S100, multi-source heterogeneous data acquisition and synchronization.

[0080] This step forms the data foundation for the entire early warning method. The system acquires data streams in parallel from various sensor modules of the data acquisition unit, specifically including: high-density 3D point cloud data describing the environment in front of the vehicle, output by the 3D perception module; pitch angle θ and roll angle φ reflecting the vehicle's real-time attitude, and vertical displacement Δz of the vehicle body reflecting the vehicle's current load state, output by the vehicle state perception module. load The data includes the vehicle's global position, real-time speed, and predicted driving path output by the positioning module. To ensure the accuracy of subsequent data fusion and analysis, all collected data are assigned a unified high-precision timestamp to achieve precise alignment of the data in the time dimension.

[0081] S200, Generation and Correction of Dynamic Vehicle Models:

[0082] This step aims to construct a dynamic model that accurately represents the vehicle's occupancy in three-dimensional space. This model is not static but evolves in real time with the vehicle's state and driving process. This step is first based on a pre-defined basic three-dimensional geometric model M of the vehicle. base Attitude rotation transformation is performed using real-time attitude angles θ and φ, combined with real-time load displacement Δz. loadA vertical translation transformation is performed to generate an initial dynamic vehicle model reflecting the current quasi-static effects. To incorporate the dynamic response of the vehicle when traveling on uneven road surfaces, this method further refines the initial dynamic vehicle model. Specifically, the method extracts the road surface elevation profile h along the predicted driving path from the 3D environment model constructed in subsequent steps. road (s), where s is the arc length along the path. This road surface profile is used as the excitation input for the vehicle suspension system dynamics model, which can be described by the following second-order differential equation:

[0083]

[0084] Where m is the equivalent mass of the suspension system, c is the equivalent damping coefficient, and k is the equivalent spring stiffness; these three are all inherent parameters of the vehicle; y(t) is derived from the road surface elevation profile h. road The excitation input at the wheel is determined by Δz(t) and the real-time vehicle speed; z(t) is the instantaneous vertical displacement response of the vehicle body obtained by solving this equation. This method calculates the instantaneous vertical displacement Δz, which varies with the path position s. d yn(s) is then applied again to the initial dynamic vehicle model via a translation transformation, ultimately resulting in a final dynamic vehicle model M that incorporates future dynamic bump responses. final (s).

[0085] S300, Construction and Enhancement of 3D Environment Models

[0086] This step is performed in parallel with the construction of the vehicle model, aiming to build an accurate model describing the passable space in front of the vehicle. A preferred implementation is to apply algorithms such as Random Sample Consensus (RANSAC) to process the 3D point cloud collected in S100, segmenting the road surface point set and fitting a high-precision road surface model, while the remaining non-road surface point cloud is identified as the locally perceived obstacle point set O. lidar To overcome the limitations of single-vehicle perception, this method also receives cooperative geometric information G broadcast by other vehicles via a communication unit. v2v And from this, the geometric model O of the obstacle detected by the external vehicle is extracted. v2v Subsequently, these external obstacle models are merged with the locally perceived obstacle point set in three-dimensional space to form a more complete global obstacle set O. total Ultimately, this global set of obstacles defines the boundaries of the passable space, forming an enhanced 3D environment model V. enhanced o S400, Predictive Interferometry.

[0087] This step is the core of risk prediction. This method uses the final dynamic vehicle model M obtained in S200, which dynamically changes with the path position s.final (s) performs a virtual simulation along the predicted driving path in both spatiotemporal dimensions. At each discrete location point s during the simulation process... k All of them perform a spatial interferometry test once, that is, determine M. final (s k The volume of ) and the global obstacle set O obtained in S300 total Does the space they occupy overlap? The mathematical condition for this can be expressed as the intersection of the three-dimensional spaces of the two geometric solids being non-empty:

[0088] S500, Decision Making and Information Distribution. If no interference is detected during the entire predicted path simulation, the path is deemed safe, and the process can return to S100 to continue looping. If interference is detected at a future location, a height restriction risk is identified. In this case, the method determines the urgency level of the risk based on the distance between the location of the first interference and the vehicle, and generates a corresponding warning command, which is sent to the human-machine interface unit (HMI) to output a graded warning signal to the driver.

[0089] Furthermore, the method also includes decisions regarding the broadcast of cooperative information. If the obstacle causing the interference is confirmed through comparison to be absent from any previously received cooperative geometric information, then the obstacle is determined to be a newly discovered risk point for the vehicle. In this case, the method extracts the precise three-dimensional geometric data of the new obstacle, encodes it into a new cooperative geometric information message, and instructs the communication unit to broadcast it to surrounding vehicles.

[0090] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for early warning of vehicle height restrictions, characterized in that, Includes the following steps: Acquire real-time status data of the vehicle, including real-time attitude information and real-time load information of the vehicle; Acquire three-dimensional environmental data in front of the vehicle; Based on the real-time status data, a dynamic vehicle model representing the three-dimensional space occupied by the vehicle is generated; Based on the three-dimensional environmental data, a three-dimensional environmental model representing the passable space in front of the vehicle is constructed; Interference analysis is performed between the dynamic vehicle model and the three-dimensional environment model to determine whether there is a risk of height restriction. Prior to the step of performing interferometric analysis, the following steps are also included: Extract the road surface contour along the path in front of the vehicle from the three-dimensional environment model; The instantaneous vertical displacement of the vehicle caused by road bumps is predicted based on the road surface profile. The dynamic vehicle model is then corrected based on the instantaneous vertical displacement. The steps for interference analysis between the dynamic vehicle model and the three-dimensional environment model are as follows: The modified dynamic vehicle model is virtually simulated along the path in front of the vehicle, and it is determined whether there is spatial overlap between the model and the boundary of the three-dimensional environment model.

2. The vehicle height restriction early warning method according to claim 1, characterized in that, The step of constructing a three-dimensional environment model representing the passable space in front of the vehicle includes: Based on the aforementioned three-dimensional environment data, an initial three-dimensional environment model is constructed; It receives cooperative geometric information broadcast via vehicle-to-vehicle communication and uses the cooperative geometric information to update the initial three-dimensional environment model to obtain the three-dimensional environment model.

3. The vehicle height restriction early warning method according to claim 2, characterized in that, When the interference analysis determines that there is a height restriction risk caused by a newly detected obstacle, the method encodes the geometric data of the obstacle into cooperative geometric information and broadcasts it through vehicle-to-vehicle communication.

4. The vehicle height restriction early warning method according to claim 1, characterized in that, The step of generating the dynamic vehicle model is as follows: For the preset vehicle base model, a rotation transformation based on the real-time attitude information and a translation transformation based on the real-time load information are applied sequentially.

5. The vehicle height restriction early warning method according to claim 1, characterized in that, The step of constructing a three-dimensional environment model representing the passable space in front of the vehicle specifically includes: The road surface is identified and fitted from the three-dimensional environmental data; The non-road surface 3D environment data is identified as obstacles, and the boundaries of the 3D environment model are defined by the obstacles.

6. The vehicle height restriction early warning method according to claim 1, characterized in that, When the interference analysis determines that there is a height restriction risk, the method executes different levels of warning prompts based on the distance between the height restriction risk and the vehicle.

7. The vehicle height restriction early warning method according to claim 1, characterized in that, The step of constructing a three-dimensional environment model representing the passable space in front of the vehicle includes: Based on the aforementioned three-dimensional environment data, an initial three-dimensional environment model is constructed; It receives cooperative geometric information broadcast via vehicle-to-vehicle communication and uses the cooperative geometric information to update the initial three-dimensional environment model to obtain the three-dimensional environment model.

8. A vehicle height restriction warning system, applied to the vehicle height restriction warning method according to any one of claims 1-7, characterized in that, include: A processor and a memory connected to the processor, the memory storing program instructions, wherein when the processor executes the program instructions, it implements the method as described in any one of claims 1 to 7.

Images