Vehicle height adjustment system and method thereof

Abstract

The present invention relates to a vehicle height control system and a method thereof. A vehicle height adjustment system according to the present invention includes a ground sensing unit that detects the condition of the ground in front of the vehicle, a height prediction unit that predicts the vehicle height based on the detected ground data, and a height adjustment unit that adjusts the suspension height based on the predicted height data.

Description

Technology Field

[0001] The present invention relates to a vehicle height adjustment system and a method for preventing damage to the vehicle undercarriage and ensuring safe driving. Background Technology

[0003] In the automotive industry, technologies are continuously evolving to ensure vehicle stability and safety under various road environments and driving conditions. In particular, technology for efficiently adjusting vehicle ground clearance plays a crucial role in preventing damage to the vehicle's undercarriage and improving ride comfort and driving safety.

[0004] Vehicle ride height adjustment is intended to maintain an appropriate distance between the vehicle's undercarriage and the ground in situations involving speed bumps, uneven roads, or obstacles, and this technology is increasingly being utilized in advanced systems such as autonomous vehicles.

[0005] Conventional vehicle ride height adjustment systems are generally based on a fixed suspension height or a mechanism that the driver can manually adjust.

[0006] However, such systems have difficulty responding immediately to sudden changes in the driving environment or obstacles, and can lead to problems such as undercarriage damage or reduced ride comfort, especially under complex road conditions. For example, when passing over speed bumps or driving over sudden uneven sections of the road, if the gap between the vehicle's undercarriage and the ground is not properly adjusted, the risk of damage to vehicle parts increases.

[0007] In addition, existing systems lack the ability to detect and analyze the ground conditions in front of the vehicle in real time to adjust the vehicle's height. Since most systems rely on limited sensor data and static adjustment rules, they have difficulty handling various types of obstacles or complex road conditions.

[0008] This causes the vehicle to fail to adapt to the new road environment, resulting in a simultaneous deterioration of safety and ride comfort. The problem to be solved

[0010] The present invention, derived from the above necessity, provides a vehicle height adjustment system and method that enables safe and efficient height adjustment even under various driving conditions by detecting the ground condition in front of the vehicle and optimizing height adjustment through a machine learning algorithm. means of solving the problem

[0012] A vehicle height adjustment system according to an embodiment of the present invention includes a ground sensing unit that detects the condition of the ground in front of the vehicle, a height prediction unit that predicts the vehicle height based on the detected ground data, and a height adjustment unit that adjusts the suspension height based on the predicted height data.

[0013] A vehicle height adjustment method according to another embodiment of the present invention includes the steps of detecting a ground condition in front of the vehicle, preprocessing the detected ground data to convert it into an analyzable data format, predicting the vehicle height based on the converted data, and adjusting the suspension height based on the predicted data.

[0014] A vehicle height adjustment system according to another embodiment of the present invention includes a data collection unit that collects ground condition and obstacle data through a camera positioned in front of the vehicle, a risk analysis unit that evaluates the risk of obstacles by analyzing data input from the data collection unit, a suspension control unit that adjusts the vehicle's suspension height in real time based on the data evaluated by the risk analysis unit, and a data learning unit that learns the data collected from the data collection unit and the risk analysis unit using a machine learning algorithm and generates a learning model for optimal height adjustment in various driving situations. Effects of the invention

[0016] According to an embodiment of the present invention, by detecting the ground condition in front of the vehicle, predicting the vehicle height through a machine learning algorithm based on the detected data, and dynamically adjusting the optimal suspension height, the effect of preventing damage to the underside of the vehicle and improving driving safety is provided.

[0017] The effects of the present invention are not limited to those mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art from the description below. Brief explanation of the drawing

[0019] FIG. 1 illustrates a vehicle height adjustment system according to one embodiment of the present invention. FIG. 2 illustrates a method for adjusting the overall height of a vehicle according to an embodiment of another aspect of the present invention. FIG. 3 illustrates a vehicle height adjustment system according to an embodiment in another aspect of the present invention. FIG. 4 is a diagram showing a signal flow for implementing a vehicle height adjustment method according to an embodiment of the present invention. Figure 5 is a diagram showing the suspension control process through risk analysis and KNN classification of the vehicle height adjustment system. Figure 6 illustrates a risk-based suspension height optimization process according to an embodiment of the present invention. Specific details for implementing the invention

[0020] The aforementioned objectives of the present invention, as well as other objectives, advantages, and features, and the methods for achieving them, will become clear from the embodiments described in detail below together with the accompanying drawings.

[0021] However, the present invention is not limited to the embodiments disclosed below but can be implemented in various different forms, and the following embodiments are provided merely to easily inform those skilled in the art of the purpose, structure, and effects of the invention, and the scope of the rights of the present invention is defined by the description in the claims.

[0022] Meanwhile, the terms used in this specification are for describing the embodiments and are not intended to limit the invention. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. As used in this specification, "comprises" and / or "comprising" do not exclude the presence or addition of one or more other components, steps, actions, and / or elements to the mentioned components, steps, actions, and / or elements.

[0024] FIG. 1 illustrates a vehicle height adjustment system according to an embodiment of the present invention.

[0025] As described above, a vehicle height adjustment system according to an embodiment of the present invention includes a ground detection unit (110) that detects the ground condition in front of the vehicle, a height prediction unit (120) that predicts the vehicle height based on the detected ground data, and a height adjustment unit (130) that adjusts the suspension height based on the predicted height data.

[0026] The present invention relates to a system that detects the condition of the ground in front of a vehicle in real time, predicts the vehicle's overall height based on the detected data, and then adjusts the suspension height to prevent damage to the vehicle's undercarriage. By identifying the ground condition, predicting the overall height using machine learning, and efficiently adjusting the suspension, the invention enhances the driving safety of the vehicle while effectively preventing damage to the undercarriage when passing through obstacles.

[0028] A ground detection unit (110) according to an embodiment of the present invention detects the condition of the ground in front of the vehicle in real time through a camera or sensor positioned in front of the vehicle, and collects height, shape, and distance data of the ground. The detected data undergoes a preprocessing step to analyze the shape of the ground and is converted into an analyzable data format. This data is used to predict the overall height of the vehicle.

[0030] The ground sensing unit (110) according to an embodiment of the present invention can detect ground height, shape, and distance data in front of the vehicle.

[0031] The ground sensing unit (110) is intended to predict the vehicle height and adjust the suspension by providing more accurate data in various driving environments, thereby enabling the safety and efficient height adjustment of the vehicle and contributing to preventing damage to the vehicle's undercarriage caused by obstacles.

[0032] The ground detection unit (110) may include a camera, lidar, or ultrasonic sensor installed at the front of the vehicle and detects the ground condition in real time. The ground detection unit (110) detects height, shape, and distance data of the ground, enabling precise analysis of the complex characteristics of the ground. Data collected through the camera is processed into 2D or 3D images, while the lidar detects depth information of the ground and the ultrasonic sensor detects proximity data to the ground. This single or multi-sensor-based data undergoes a preprocessing process before being transmitted to the overall height prediction unit (120) and is converted into an analyzable data format.

[0033] The ground detection unit (110) can provide high reliability in various driving environments by preprocessing the detected data. The ground height can be used to measure the height of obstacles and analyze the possibility of collision with the underside of the vehicle. Shape data is used to optimize the vehicle height by identifying the shape of obstacles such as speed bumps. Distance data contributes to evaluating the speed at which the vehicle approaches obstacles and the condition of the ground, thereby increasing the prediction accuracy for height adjustment.

[0034] In a specific embodiment, when a vehicle travels on a road with a speed bump, the ground detection unit (110) detects the height and shape of the speed bump and collects data. The detected data determines the size of the speed bump through ground height information and analyzes its width and shape through shape data. The distance data provides an accurate distance between the vehicle and the obstacle and is used to predict the overall height adjustment before the vehicle approaches. For example, if the detection unit detects that a vehicle is approaching a speed bump with a height of 10 cm at a distance of 3 m, this data is transmitted to the overall height prediction unit (120) to allow for preparation of an appropriate overall height adjustment.

[0035] Accordingly, according to an embodiment of the present invention, the accuracy of detecting the ground condition in front of the vehicle is improved, and by detecting ground height, shape, and distance data in real time, driving safety and ride comfort are provided simultaneously, and stable overall height adjustment is possible even in various driving environments.

[0037] According to an embodiment of the present invention, the ground sensing unit (110) preprocesses the detected data and converts it into an analyzable data format.

[0038] This is because the detected raw data may not be directly suitable for analysis, so a preprocessing step improves data quality and enables the analysis and prediction processes to be performed more efficiently.

[0039] The ground detection unit (110) includes a preprocessing step that detects the ground condition in front of the vehicle and converts the data into an analyzable format. This preprocessing process removes noise from the raw data, extracts necessary data points, and normalizes the data to convert it into a format suitable for analysis and prediction. For example, image data collected through a camera is processed at the pixel level to extract height and shape information of the ground, and depth data detected through a lidar is converted into a 3D format. This preprocessing process can reduce the complexity of the data and increase the efficiency of ground condition analysis.

[0040] The preprocessing process includes data normalization, filtering, and data reconstruction. In the normalization process, the scale of the detected data is adjusted to convert it into a value suitable for analysis and learning. The filtering process removes unnecessary noise from the detected data and preserves key information for accurate analysis. The data reconstruction process rearranges the raw data into an analyzable structure and separates the height, shape, and distance information of the ground to transmit to the height prediction unit (120) and the suspension adjustment unit.

[0041] In a specific embodiment, when a vehicle detects speed bump data through a camera while driving, the detected image data may contain various environmental noises in its raw state, and if used directly for analysis, there is a high probability of errors occurring. The ground detection unit (110) preprocesses this data to accurately extract the height and shape of the speed bump. For example, unnecessary background data is removed from the camera image, the contour of the speed bump is filtered, and the ground height data is converted from 2D to 3D to create an analyzable format. When this data is transmitted to the overall height prediction unit (120), it can be used as base data for accurate prediction and suspension adjustment.

[0043] The overall height prediction unit (120) according to an embodiment of the present invention predicts changes in the overall height of a vehicle based on data provided by the ground detection unit (110), and can predict how the overall height of the vehicle will change depending on the condition of the ground using a machine learning algorithm. The learned model is designed to reflect the relationship between the ground condition and the overall height of the vehicle in various driving environments, thereby generating more accurate overall height prediction values ​​in real time, and the predicted data can be used as a standard for suspension adjustment thereafter.

[0044] The overall height prediction unit (120) can learn the relationship between the ground condition and the overall height of the vehicle in various driving environments and calculate an accurate overall height prediction value in real time to dynamically adjust the suspension height, thereby preventing damage to the underside of the vehicle and improving driving safety.

[0045] The overall height prediction unit (120) receives data provided by the ground detection unit (110) as input, analyzes it, and predicts changes in the overall height of the vehicle. A machine learning algorithm is utilized, and the algorithm analyzes the correlation between the ground condition and the overall height of the vehicle based on various driving data and a learning model. The overall height prediction unit (120) preprocesses the input data to extract necessary features and applies them to a learned model to calculate changes in the overall height of the vehicle in real time. This structure enables precise adjustment of the overall height according to the complexity of the ground condition and various driving situations.

[0046] Machine learning algorithms learn the data necessary to optimize vehicle height based on the ground conditions in front of the vehicle. During the learning process, various input elements including ground height, shape, and distance data are used, and the algorithm models the pattern of change in vehicle height based on this. The trained model can perform the function of dynamically predicting vehicle height by considering various factors such as vehicle speed, obstacle type, and size.

[0047] In a specific embodiment, when a vehicle approaches a speed bump on a road, the ground detection unit (110) detects the height and distance data of the speed bump and transmits this to the overall height prediction unit (120). The overall height prediction unit (120) uses a learned machine learning model to predict a change in the vehicle's overall height based on the height of the speed bump. For example, if the detected height of the speed bump is 12 cm and the vehicle speed is 40 km / h, the overall height prediction unit (120) calculates a result indicating that the suspension must be adjusted to 15 cm to prevent damage to the vehicle's undercarriage. This result is transmitted to the suspension control unit to adjust the vehicle's overall height in real time.

[0048] Accordingly, according to an embodiment of the present invention, changes in vehicle height can be precisely predicted based on the condition of the ground in front of the vehicle by utilizing a machine learning algorithm, thereby preventing damage to the underside of the vehicle, providing driving stability and ride comfort in various driving environments, and improving the efficiency of the vehicle height adjustment system.

[0050] The machine learning algorithm according to an embodiment of the present invention may include a K-Nearest Neighbors (KNN) algorithm. The KNN algorithm predicts changes in overall height by comparing similar ground conditions and vehicle driving data based on detected data, which provides high accuracy even with a simple structure and enables real-time adjustment of overall height.

[0051] The KNN algorithm included in the overall height prediction unit (120) is a data-based analysis method that performs prediction based on the similarity between the detected data and the existing training data. The KNN algorithm positions the input data in a multidimensional space and determines the change in overall height based on the K closest data points. This algorithm is a simple yet efficient prediction model that provides high prediction accuracy for input data similar to the distribution of the training data. In addition, the KNN algorithm can adapt to various driving environments through continuous updates of the trained model.

[0052] The KNN algorithm utilizes distance metrics to measure the similarity between data points. Typically, it calculates the distance between input data and existing training data using Euclidean or Manhattan distances, and assigns weights to the K nearest data points. This approach reflects the similarity between the input data and existing data, enabling real-time prediction of total height changes. Furthermore, since the KNN algorithm operates based on existing data during the computation process, it facilitates data training and application in new environments.

[0053] In a specific embodiment, when a vehicle travels on a road with speed bumps, the ground detection unit (110) detects the height and distance data of the speed bumps and transmits them to the overall height prediction unit (120). Then, the KNN algorithm included in the overall height prediction unit (120) compares the detected data with existing training data. For example, if the height of a speed bump similar to the training data was 10 cm and the overall height was adjusted to 12 cm in that situation, the KNN algorithm performs a similar overall height adjustment in the current situation based on this. Through this, the suspension height can be optimized so that the vehicle can safely pass over the speed bumps.

[0055] The overall height adjustment unit (130) according to an embodiment of the present invention adjusts the suspension height of the vehicle in real time based on data predicted by the overall height prediction unit (120), and dynamically changes the suspension height to maintain an optimal distance between the vehicle's undercarriage and the ground. Through this, damage to the undercarriage is prevented when the vehicle passes through an obstacle, and driving stability according to the ground conditions can be secured.

[0057] The overall height adjustment unit (130) according to an embodiment of the present invention can optimize the suspension height using an algorithm.

[0058] The algorithm is designed to solve pathfinding and optimization problems by calculating the optimal path required for vehicle ride height adjustment and optimizing the suspension height based on the time taken and path efficiency. This prevents damage to the vehicle's undercarriage and enhances the accuracy and reliability of ride height adjustment in various driving environments.

[0059] The height adjustment unit (130) dynamically adjusts the suspension height of the vehicle using an algorithm. The algorithm calculates the most suitable height adjustment value through state-space search, and for this purpose, uses an evaluation function f(x)=g(x)+h(x). Here, g(x) represents the cost of movement to the current state, and h(x) represents the estimated cost (heuristic value) to the target state. This evaluation function searches for an optimal path by considering the vehicle's current state and ground conditions, and enables the suspension height to be adjusted based on this. In addition, this algorithm calculates the weight of the path and can perform real-time height adjustment according to the vehicle's driving conditions.

[0060] The algorithm comprehensively analyzes various data, such as distance, speed, and obstacle heights, to determine the optimal suspension height in a vehicle driving environment. During the pathfinding process, the algorithm evaluates all possible routes and applies priorities to find the optimal route in the shortest possible time.

[0061] Furthermore, by utilizing a weighted algorithm, faster and more efficient overall height adjustment is possible by adjusting weights in the form f(x)=g(x)+αh(x)(α>1) in specific situations to amplify the influence of heuristics.

[0062] In a specific embodiment, when a vehicle is driving on a road with complex obstacles, the vehicle height adjustment unit (130) uses an algorithm to calculate the optimal distance between the vehicle and the ground. For example, when the height of the detected obstacle is 15 cm and the current speed of the vehicle is 50 km / h, the algorithm can adjust the suspension to 18 cm based on this data to prevent a collision between the underside of the vehicle and the obstacle. Additionally, if the height of the obstacle decreases, the algorithm readjusts the suspension height to reflect this and maintain driving stability.

[0063] Accordingly, according to an embodiment of the present invention, by optimizing the suspension height of a vehicle using an algorithm, damage to the vehicle's undercarriage is prevented, and stable overall height adjustment is enabled even in various driving environments. The algorithm efficiently performs overall height adjustment through path search and cost calculation, thereby providing both driving stability and ride comfort.

[0065] The algorithm according to an embodiment of the present invention may be configured to apply distance weights for calculating path costs.

[0066] Distance weights are factors used to calculate path costs more precisely based on the characteristics of various ground conditions and obstacles encountered while driving, thereby improving the efficiency and accuracy of the algorithm. By utilizing distance weights, vehicle height adjustment can be performed more quickly and accurately.

[0067] The algorithm uses an evaluation function f(x) = g(x) + h(x) during pathfinding, where g(x) represents the actual cost to the current state and h(x) represents the estimated cost (heuristic value) to the target state. Distance weights are applied by assigning weights to g(x) and h(x) in this evaluation function to calculate the path cost. For example, if a vehicle must pass through a high obstacle, the weight of that path is increased to guide the algorithm to select a more suitable route. This can improve the precision of path cost calculation and enhance the efficiency of obstacle avoidance and ground clearance adjustment.

[0068] Distance weights are calculated by comprehensively considering factors such as the physical distance between the vehicle's current location and the target location, the complexity of the ground conditions, and the height and shape of obstacles. For example, given a path with speed bumps and a flat path, a higher weight is applied to the path with speed bumps, and a lower weight is applied to the flat path. This allows the algorithm to search for a cost-effective path and optimize the vehicle's ground clearance. Additionally, weights can be changed in real time, which supports the vehicle in adapting to dynamic environmental changes while driving.

[0069] In a specific embodiment, in a situation where a vehicle must select one of two paths, the first path contains an obstacle with a height of 15 cm, including a speed bump, while the second path consists of a flat road. The algorithm collects information on the distance and obstacles of the two paths, and then assigns a weight to the path containing the speed bump to increase the cost. For example, if the weight of the first path is set to 1.5 and the weight of the second path is set to 1, the algorithm prioritizes the second path and performs height adjustment. This process maintains the safety and efficiency of the vehicle and prevents energy consumption caused by unnecessary height adjustment.

[0070] Accordingly, according to an embodiment of the present invention, the algorithm can efficiently perform vehicle height adjustment by applying distance weights to precisely calculate the path cost. In addition, by utilizing distance weights, the vehicle can adjust its height more quickly and accurately when encountering obstacles or complex ground conditions, thereby preventing damage to the vehicle's undercarriage and improving the performance of the height adjustment system.

[0072] As such, in the vehicle height adjustment system according to an embodiment of the present invention, for example, when a vehicle is driving on a road with a speed bump, the ground detection unit (110) detects the height and distance of the speed bump in front of the vehicle and transmits this to the height prediction unit (120). The height prediction unit (120) calculates an appropriate height for passing over the speed bump using a learned model. The height adjustment unit (130) immediately adjusts the suspension height based on this data to prevent damage to the underside of the vehicle when passing over the speed bump. This operation is performed in real time to maintain vehicle driving safety and improve ride comfort.

[0073] Accordingly, the vehicle height adjustment system according to an embodiment of the present invention prevents damage to the vehicle's undercarriage when the vehicle travels on various ground conditions and simultaneously provides driving stability and ride comfort. Through this, it maintains the safety of the vehicle and reduces structural damage caused by obstacles or ground irregularities, while simultaneously enabling efficient height control by adjusting the suspension height in real time.

[0075] FIG. 2 illustrates a method for adjusting vehicle height according to an embodiment of another aspect of the present invention. In this embodiment, detailed descriptions of components that are partially duplicated in the above-described embodiment are omitted.

[0076] As described above, the method may include a step of detecting the ground condition in front of the vehicle (S710), a step of preprocessing the detected ground data and converting it into an analyzable data format (S720), a step of predicting the vehicle height based on the converted data (S730), and a step of adjusting the suspension height based on the predicted data (S740).

[0077] This relates to a method for detecting the condition of the ground in front of a vehicle, preprocessing the detected data to convert it into an analyzable data format, and then predicting the vehicle height based on this to adjust the suspension height. By processing the data step by step and adjusting the vehicle height in real time, it is possible to prevent damage to the underside of the vehicle and maintain driving stability in various driving environments.

[0079] The step of detecting the ground condition (S710) according to an embodiment of the present invention involves detecting the ground condition in front of the vehicle in real time using a ground detection unit, such as a camera, lidar, or ultrasonic sensor installed in front of the vehicle. In this step, height, shape, and distance data of the ground are collected, and the detected raw data is used as a basic input value for predicting the vehicle's overall height. This detection step can be used to more accurately determine the condition between the vehicle's undercarriage and the ground in various driving environments.

[0081] The step (S720) of preprocessing detected ground data according to an embodiment of the present invention improves the quality of raw data and converts the data into a format suitable for analysis and prediction, thereby removing noise from the data, filtering necessary data points, and normalizing the data to enable precise overall height prediction. For example, background data is removed from camera image data, and the contour and height data of the ground are extracted and converted into normalized analysis data.

[0083] The step (S730) of predicting the vehicle height based on converted data according to an embodiment of the present invention involves applying preprocessed data to a machine learning algorithm to predict changes in the vehicle height, wherein the machine learning model learns the correlation between the ground condition and the vehicle height and calculates the optimal vehicle height according to the ground condition. The predicted height data is used to adjust the suspension height in a subsequent step. The above step provides real-time data analysis and prediction functions, and enables rapid and accurate height adjustment in various driving environments.

[0084] The step (S730) of predicting the overall height according to an embodiment of the present invention may include a process of predicting a change in the vehicle's overall height according to ground conditions using a machine learning algorithm. This process calculates a predicted value for optimizing the vehicle's overall height based on ground height, shape, and distance data, thereby increasing the accuracy of the overall height adjustment and enabling real-time response.

[0085] A machine learning algorithm according to an embodiment of the present invention may include a Kin Neighbors (KNN) algorithm. This predicts changes in vehicle height based on the most similar data by comparing ground data in front of the vehicle with existing training data. The KNN algorithm enables real-time height prediction with high accuracy by evaluating the similarity between input data and existing training data.

[0087] The step (S740) of adjusting the suspension height based on predicted data according to an embodiment of the present invention optimizes the distance between the vehicle's undercarriage and the ground by adjusting the suspension height in real time based on the predicted result of the vehicle's overall height, and the overall height adjustment unit dynamically controls the suspension height to prevent damage to the vehicle's undercarriage and ensure driving stability when passing through obstacles. The result of adjusting the suspension height in real time improves the physical safety and ride comfort of the vehicle.

[0088] The step (S740) of adjusting the suspension height according to an embodiment of the present invention may include a process of optimizing the suspension height using an algorithm.

[0089] The step of adjusting the suspension height (S740) utilizes an algorithm to perform path search and optimization, and dynamically adjusts the suspension height of the vehicle. The algorithm calculates path costs and estimated costs to search for an optimal path for adjusting the vehicle's overall height, and supports safe driving by reflecting this in real time.

[0090] An algorithm according to an embodiment of the present invention may include a process of applying distance weights for calculating path costs. The algorithm applies distance weights when calculating path costs to comprehensively consider various factors such as ground conditions, vehicle speed, and the height of obstacles. The distance weights increase the precision of path cost calculation and optimize the suspension height so that the vehicle can safely pass through obstacles and ensure driving stability.

[0092] As described above, in the vehicle height adjustment method according to an embodiment of the present invention, when a vehicle is driving on a road with a speed bump, the height and distance of the speed bump in front of the vehicle are detected in the first step. The detected data is preprocessed in the second step and converted into an analyzable data format including height and shape information of the speed bump. The converted data is analyzed through a machine learning algorithm in the third step to predict a suspension height for optimizing the vehicle's height. In the fourth step, the suspension height is adjusted to, for example, 15 cm, enabling the vehicle to safely pass over the speed bump.

[0093] Accordingly, the vehicle height adjustment method according to the embodiment of the present invention detects the ground conditions in front of the vehicle in real time and converts the data into an analyzable format, thereby accurately predicting the vehicle height and dynamically adjusting the suspension height, thereby preventing damage to the underside of the vehicle and providing driving stability and ride comfort in various driving environments.

[0095] FIG. 3 illustrates a vehicle height adjustment system according to an embodiment of another aspect of the present invention. In this embodiment, detailed descriptions of components that are partially duplicated in the above-described embodiment are omitted.

[0096] As described above, a vehicle height adjustment system according to an embodiment of the present invention includes a data collection unit (210) that collects ground condition and obstacle data through a camera positioned in front of the vehicle, a risk analysis unit (220) that analyzes data input from the data collection unit (210) to evaluate the risk of obstacles, a suspension control unit (230) that adjusts the suspension height of the vehicle in real time based on the data evaluated by the risk analysis unit (220), and a data learning unit (240) that learns the data collected from the data collection unit (210) and the risk analysis unit (220) using a machine learning algorithm and generates a learning model for optimal height adjustment in various driving situations.

[0097] This involves detecting the ground conditions and obstacles in front of the vehicle, analyzing the detected data to adjust the vehicle's suspension height in real time, and learning an optimal height adjustment model through machine learning. By integrally performing data collection, risk analysis, suspension control, and the creation of a learning model, it prevents damage to the vehicle's undercarriage and provides driving stability in various driving environments.

[0099] The data collection unit (210) according to an embodiment of the present invention collects ground condition and obstacle data in real time through a camera positioned in front of a vehicle, and the collected data includes the height, shape, and distance of the ground and the size and location of obstacles, and is used as initial input data for the system. The data collection unit (210) is designed to operate stably in various driving environments and provides data capable of accurately detecting ground conditions.

[0100] The data collection unit (210) according to an embodiment of the present invention is configured to detect ground height, shape, and distance data in front of the vehicle, accurately identifies the condition of the ground in front, and collects data in real time and transmits it to other components of the system to prevent damage to the underside of the vehicle and ensure driving safety.

[0101] The data collection unit (210) includes a camera, lidar, or ultrasonic sensor positioned at the front of the vehicle and is configured to detect ground height, shape, and distance data in real time. The camera collects 2D or 3D image data to precisely measure the contour and height of the ground, while the lidar provides depth data to accurately analyze the distance from the ground and the shape of obstacles. The ultrasonic sensor is suitable for detecting short-range data and is used to measure the distance between the vehicle and the ground. These sensors operate complementarily to provide accurate data in various driving environments.

[0102] The data collection unit (210) preprocesses the detected data and converts it into an analyzable format, and transmits it to the risk analysis unit (220), suspension control unit (230), and data learning unit (240). For example, raw image data collected by a camera removes noise and precisely calculates the ground height and shape through contour extraction and filtering, and the depth data from the lidar is used to analyze how close the vehicle is to the ground. This data contributes to increasing the accuracy of the overall height prediction and suspension adjustment processes.

[0103] In a specific embodiment, when a vehicle is traveling on a road with a speed bump, the data collection unit (210) detects the height and shape of the speed bump in front of the vehicle and collects data. A camera records the height and width of the speed bump as image data, and a lidar precisely measures the height of the speed bump and the distance between the vehicle and the speed bump. An ultrasonic sensor detects the shortest distance between the vehicle and the speed bump to supplement the data. By collecting and processing this data in real time and transmitting it to other components, the suspension height is adjusted so that the vehicle can safely pass over the speed bump.

[0104] Accordingly, according to an embodiment of the present invention, the data collection unit (210) accurately detects the ground height, shape, and distance data in front of the vehicle in real time, thereby supporting the vehicle height adjustment system to operate stably in various driving environments. Through this, damage to the underside of the vehicle is prevented, driving safety is maintained, and data-based height adjustment is enabled, thereby improving the performance and reliability of the system.

[0106] The risk analysis unit (220) according to an embodiment of the present invention evaluates the risk of an obstacle by analyzing data transmitted from the data collection unit (210), and the risk is graded by considering the ground condition and the size, shape, distance, etc. of the obstacle. For example, the risk is classified into grades such as F and B, and the higher the risk, the more immediately the vehicle height adjustment is designed to be performed. The risk analysis unit (220) evaluates the situation faced by the vehicle in real time and provides necessary information to the suspension control unit (230).

[0108] The risk analysis unit (220) according to an embodiment of the present invention is configured to evaluate the possibility of damage to the underside of the vehicle based on ground conditions and obstacle data.

[0109] The risk analysis unit (220) is configured to receive ground height, shape, and distance data and obstacle data provided by the data collection unit (210) as input, analyze them, and evaluate the possibility of collision with the underside of the vehicle. It calculates the risk level of the obstacle based on the characteristics of the data and comprehensively analyzes the ground condition, obstacle size, distance, shape, etc. The risk level is classified into grades such as F and B and is used to determine whether vehicle height adjustment is necessary and the urgency thereof. For example, the design ensures that immediate suspension height adjustment is performed for obstacles with higher risk levels. The results of this analysis are transmitted to the suspension control unit (230) and the data learning unit (240) and reflected in the vehicle height adjustment.

[0110] The risk analysis unit (220) preprocesses data and, based on this, sets specific criteria to evaluate the possibility of damage to the vehicle's undercarriage. If the gap between the obstacle height and the vehicle's overall height decreases below a specific threshold value, the risk analysis unit (220) classifies this as a high risk and instructs the suspension height to be adjusted immediately. Conversely, if the gap is sufficiently secured, it is evaluated as a low risk and the basic overall height is maintained. This analysis is performed by comprehensively considering the ground conditions, the vehicle's current speed, and whether the obstacle is moving.

[0111] In a specific embodiment, when a vehicle approaches a speed bump 15 cm high, the data collection unit (210) detects the height and distance of the speed bump and transmits this to the risk analysis unit (220), and the risk analysis unit (220) calculates the possibility that the gap between the speed bump and the underside of the vehicle will narrow. For example, if the vehicle height is currently 13 cm, the risk analysis unit (220) evaluates that there is a high risk of damage to the underside and classifies it as Grade F. This result is immediately transmitted to the suspension control unit (230), and the vehicle height is adjusted to 17 cm. This risk analysis ensures that the vehicle maintains a safe gap when passing over an obstacle.

[0112] By doing so, the possibility of undercarriage damage can be more accurately assessed based on the ground conditions and obstacles in front of the vehicle, allowing the vehicle height to be adjusted in real time. This prevents undercarriage damage caused by obstacles, provides driving safety and ride comfort simultaneously, and increases the reliability and efficiency of the system to operate stably in various driving environments.

[0114] The suspension control unit (230) adjusts the suspension height of the vehicle in real time based on data provided by the risk analysis unit (220), and dynamically controls the suspension to maintain an optimal distance between the vehicle's undercarriage and the ground, thereby preventing damage to the vehicle's undercarriage that may occur when passing through obstacles and enhancing driving stability.

[0115] The suspension control unit (230) is configured to receive risk data transmitted from the risk analysis unit (220) as input and dynamically adjust the suspension height of the vehicle. When the risk is assessed as high, the suspension control unit (230) increases the suspension height to prevent damage to the vehicle's undercarriage, and when the risk is assessed as low, it maintains or lowers the basic overall height to improve the stability and efficiency of the vehicle's driving. This adjusts the suspension height by comprehensively considering the vehicle's current speed, the size and distance of obstacles, and ground conditions. The suspension control unit (230) provides real-time data processing and a fast response speed, thereby ensuring the driving safety of the vehicle.

[0116] The suspension control unit (230) adjusts the optimal height by considering the load distribution and driving stability of the vehicle. In this process, the suspension height adjustment is performed in stages based on the results of the risk analysis and is smoothly synchronized with the physical movement of the vehicle. For example, if the vehicle encounters a sudden obstacle, the suspension control unit (230) operates immediately to prevent the underside of the vehicle from colliding with the obstacle.

[0117] In a specific embodiment, when a vehicle is driving on a road with a speed bump, the risk analysis unit (220) transmits data to the suspension control unit (230) that has been evaluated as having a high risk of grade F based on the height of the speed bump. Based on this data, the suspension control unit (230) adjusts the vehicle's current suspension height from 15 cm to 18 cm to secure a safe distance between the vehicle's undercarriage and the speed bump. After the vehicle passes the speed bump, the suspension control unit (230) readjusts the suspension height to the basic overall height to maintain a normal driving state. This real-time adjustment ensures both the stability of the vehicle and protection of the undercarriage.

[0118] Accordingly, the suspension control unit (230) adjusts the vehicle suspension height in real time based on the data of the risk analysis unit (220), thereby preventing damage to the vehicle's undercarriage and providing driving stability, and minimizes the risk caused by obstacles, thereby enhancing the safety of the vehicle under various driving conditions.

[0120] The suspension control unit (230) according to an embodiment of the present invention may be configured to optimize the suspension height using an algorithm.

[0121] The algorithm is capable of efficiently performing pathfinding and cost calculation, and plays a crucial role in ensuring driving stability and preventing undercarriage damage by optimizing suspension height during vehicle ride height adjustment. This enables the vehicle to operate safely in various driving environments.

[0122] The suspension control unit (230) uses an algorithm to analyze data such as ground conditions, the size and distance of obstacles, and vehicle speed, and calculates the optimal suspension height based on this. The algorithm uses f(x)=g(x)+h(x) as an evaluation function to calculate the path cost, where g(x) represents the actual cost to the current state and h(x) represents the estimated cost (heuristic value) to the target state. The suspension control unit (230) calculates the optimal overall height value through this evaluation function and adjusts the suspension height in real time based on this. In addition, the algorithm searches for an optimal path that can increase efficiency and maintain vehicle driving stability by applying various weights during the path search process.

[0123] The algorithm searches for all possible paths required for suspension height adjustment and then selects the minimum cost path to adjust the vehicle's overall height. In this process, the vehicle's load, ground conditions, and the height and shape of obstacles are taken into account in an integrated manner, thereby maintaining both the vehicle's safety and ride comfort. The suspension control unit (230) operates in real time to enable the vehicle to drive stably even when passing through obstacles or in sections with sudden changes in ground.

[0124] In a specific embodiment, when a vehicle is driving on a road with a speed bump, the data collection unit (210) detects the height and distance of the speed bump and transmits this to the suspension control unit (230), and the suspension control unit (230) calculates the optimal suspension height using an algorithm. For example, if the height of the speed bump is 10 cm and the current height of the vehicle is 8 cm, the algorithm calculates this and adjusts the suspension height to 12 cm. This optimization maintains a safe distance between the vehicle's undercarriage and the obstacle, and enables the vehicle to pass over the speed bump stably.

[0125] Accordingly, the suspension control unit (230) according to the embodiment of the present invention uses an algorithm to optimize the vehicle suspension height, thereby preventing damage to the vehicle's undercarriage and providing driving stability and ride comfort in various driving environments.

[0127] The algorithm according to an embodiment of the present invention may be configured to apply distance weights for calculating path costs.

[0128] Distance weights are used to precisely calculate path costs by comprehensively considering various factors such as the size and location of obstacles encountered while driving, ground conditions, and vehicle speed. This increases the efficiency of the algorithm and optimizes the vehicle suspension height to prevent damage to the vehicle's undercarriage and enhance driving safety.

[0129] The algorithm performs path search and cost calculation in the suspension control unit (230), and distance weights are used as an important factor in adjusting the path evaluation function during this process. The path evaluation function is defined as f(x)=g(x)+h(x), where g(x) represents the actual travel cost up to the present and h(x) represents the estimated cost to the target state. Distance weights influence the calculation of g(x) and h(x) in this function, dynamically adjusting the path cost according to driving conditions such as the height and distance of obstacles and vehicle speed. These dynamic weights support the calculation of the optimal suspension height in real time when the vehicle encounters obstacles or complex ground conditions.

[0130] Distance weights are applied by comprehensively analyzing various data, such as the physical distance between the vehicle and obstacles, as well as the obstacles' height and shape, and vehicle speed. For example, paths with high obstacles are assigned higher distance weights, prompting the algorithm to avoid them or adjust the suspension height more significantly. Conversely, flat paths are assigned lower distance weights to maintain the standard suspension height. This approach helps the algorithm find efficient routes and perform precise adjustments to the vehicle's ride height.

[0131] In a specific embodiment, when a vehicle must choose between a speed bump and a flat road, the path with the speed bump is assigned a higher cost and weight based on height and distance, while the flat road is assigned a lower cost and weight. For example, if the height of the speed bump is 15 cm and the vehicle's current speed is 40 km / h, the algorithm assigns a weight of 1.5 to the speed bump path and a weight of 1.0 to the flat path, thereby encouraging the vehicle to prioritize the flat road. This path selection supports maintaining safety and driving efficiency through vehicle height adjustment.

[0132] Accordingly, according to an embodiment of the present invention, by applying distance weights in an algorithm to precisely calculate the path cost, the vehicle suspension height is optimized, and stable operation is enabled even in various driving environments. The distance weights more precisely reflect obstacles and ground conditions during the path search process, thereby preventing damage to the vehicle's undercarriage and simultaneously providing driving safety and ride comfort.

[0134] The data learning unit (240) according to an embodiment of the present invention learns data collected from the data collection unit (210) and the risk analysis unit (220) through a machine learning algorithm and generates a learning model for optimal overall height adjustment in various driving situations. The learned model is continuously updated, which supports the system in adapting to new environments and data to operate with higher reliability. The data learning unit (240) improves the predicted value in real time based on the learning data and enhances the driving stability of the vehicle and the accuracy of the system.

[0135] The data learning unit (240) according to an embodiment of the present invention is configured to learn data collected through a machine learning algorithm and update the learning data so that the vehicle height can be adjusted according to the type of obstacle.

[0136] The data learning unit (240) adapts to various driving environments and new types of obstacles through continuous learning, thereby improving the accuracy and reliability of vehicle height adjustment.

[0137] The data learning unit (240) receives ground condition and obstacle data provided by the data collection unit (210) and the risk analysis unit (220) as input and generates and updates training data using a machine learning algorithm. In this process, the learning unit analyzes key characteristics such as the height, shape, and distance of obstacles, and based on this, trains a model for adjusting the vehicle height. The machine learning algorithm continuously learns new data to maintain the accuracy of the training model and supports stable operation under various driving conditions. The trained data is used in the process of real-time height prediction and suspension control, and increases the adaptability and efficiency of the system.

[0138] The data learning unit (240) models rules and patterns for adjusting the vehicle height according to the type of obstacle based on the learned data. The types of obstacles are classified based on height and shape, and generate height adjustment values ​​suitable for each type. For example, speed bumps and potholes require different adjustment values. The data learning unit (240) learns these type-specific patterns and applies them to the height adjustment process in real time to prevent damage to the underside of the vehicle.

[0139] In a specific embodiment, when a vehicle repeatedly encounters obstacles of various heights located on the road, the data learning unit (240) learns the obstacle height and shape data detected by the data collection unit (210) and calculates the overall height adjustment value for each type of obstacle. For example, if the learned model generates a rule that the suspension should be adjusted to 15 cm on a speed bump 12 cm high and to 7 cm on a road with a 5 cm deep pothole, the data learning unit (240) updates this so that it can respond quickly and accurately in similar situations thereafter. In addition, when a new type of obstacle appears, the learning unit learns it, updates the model, and maintains the accuracy of the overall height adjustment.

[0140] Accordingly, the data learning unit (240) according to the embodiment of the present invention utilizes a machine learning algorithm to continuously learn and update vehicle height adjustment data according to obstacle types, thereby preventing damage to the vehicle undercarriage and improving the reliability and adaptability of the height adjustment system in various driving environments. The learned data supports the system to operate stably even in new environments and provides higher accuracy and efficiency in the real-time height adjustment process.

[0141] The machine learning algorithm according to an embodiment of the present invention includes a Kin Neighbors (KNN) algorithm to learn obstacle data and utilize it for vehicle height adjustment. The KNN algorithm calculates a height adjustment value based on the most similar data by comparing input data with existing training data, and provides high accuracy while being simple, making it suitable for real-time learning and data application.

[0142] The KNN algorithm learns patterns for adjusting vehicle height by analyzing the type of obstacle and ground condition in the data learning unit (240). This involves comparing input data with existing training data in a multidimensional space and selecting the K closest data points. The selected data is used to calculate the characteristics of new obstacles and height adjustment values ​​by applying weights. The KNN algorithm calculates the optimal value for adjusting vehicle height based on ground height, shape, and distance data collected from the front of the vehicle, and accuracy improves as the amount of training data increases. Additionally, it enhances the adaptability of the system by learning new data and provides stable performance in various driving environments.

[0143] The KNN algorithm calculates similarity using distance metrics between input and training data. Typically, Euclidean or Manhattan distances are used to measure the similarity between data points and select the K closest data points. Due to its simple structure, the algorithm is suitable for real-time data processing and can efficiently model overall height adjustment patterns based on obstacle height, shape, and distance data. Furthermore, the training model can be continuously updated by adding new training data to the KNN algorithm.

[0144] In a specific embodiment, when a vehicle is in a road situation where it must drive over several obstacles of different heights in succession, the data learning unit (240) learns the height and shape of the first obstacle and, based on this, uses a KNN algorithm to predict the overall height adjustment value of the subsequent obstacle. For example, after learning data for a speed bump with a height of 12 cm, if a new obstacle with a height of 10 cm is detected, the KNN algorithm suggests adjusting the suspension height to 13 cm based on the existing data. This process ensures that the learning model is continuously updated whenever new data is added, thereby maintaining the accuracy of the vehicle's overall height adjustment.

[0145] Accordingly, the data learning unit (240) according to the embodiment of the present invention efficiently learns obstacle data by utilizing the KNN algorithm and provides accurate values ​​for vehicle height adjustment, thereby preventing damage to the vehicle undercarriage and maintaining high reliability and adaptability in various driving environments.

[0146] This KNN algorithm is suitable for real-time data processing and learning due to its simple and efficient structure, and can enhance vehicle safety and driving stability by generating optimized height adjustment values ​​for each obstacle type.

[0148] In this manner, in the vehicle height adjustment system according to an embodiment of the present invention, when a vehicle is driving on a road with speed bumps, the data collection unit (210) detects the height and distance data of the speed bumps in real time and transmits them to the risk analysis unit (220). The risk analysis unit (220) evaluates the size of the speed bumps and classifies them as high risk. Based on this data, the suspension control unit (230) adjusts the suspension height so that the vehicle's undercarriage is not damaged by the speed bumps. At the same time, the data learning unit learns this data and generates a learning model that can adjust the suspension height more quickly and accurately in similar situations thereafter.

[0149] According to an embodiment of the present invention, the ground conditions and obstacles in front of the vehicle are detected more accurately, and the vehicle height is adjusted in real time by analyzing the data, thereby preventing damage to the vehicle's undercarriage and providing driving stability and ride comfort.

[0151] FIG. 4 is a diagram showing a signal flow for implementing a vehicle height adjustment method according to an embodiment of the present invention.

[0152] (1) Acquire image information from the camera

[0153] Video data including road surface conditions and obstacle information is acquired in real time through a camera of a data collection unit installed at the front of the vehicle. The camera captures all situations that may occur along the vehicle's driving path, and this is used as basic data for subsequent data analysis and learning.

[0154] (2) Ground detection

[0155] Based on image information acquired from the data collection unit, the ground detection unit analyzes and detects height, shape, and distance data of the ground. In this process, key characteristics of the ground are extracted to enable the evaluation of the distance from the vehicle's undercarriage and the road condition.

[0156] (3) Image preprocessing

[0157] An image preprocessing process is performed to convert raw data collected from the ground detection unit into an analyzable form. In this process, noise is removed from the image data, and data is generated to clearly recognize obstacles and road shapes.

[0158] (4) Extract detected ground information

[0159] Based on preprocessed data, the ground sensing unit extracts ground information and height changes according to specific ground features (e.g., hills, speed bumps, etc.). This is used to precisely identify ground characteristics and improve the accuracy of vehicle height adjustment.

[0160] (5) Prediction of the previous height

[0161] The overall height prediction unit predicts the vehicle's overall height difference by utilizing detected ground information. In this process, a machine learning algorithm is applied to learn changes in the overall height difference according to road conditions and supports the prediction of the optimal overall height in various driving situations.

[0162] (6) Calculate weights

[0163] Based on the predicted total height data, weights to be used in the suspension control unit are calculated. The weights comprehensively reflect data such as driving conditions, ground conditions, and obstacle heights, and are used as criteria to optimize the suspension height.

[0164] (7) Previous training data (KNN)

[0165] Utilizing the Kinetic Neighbors (KNN), a machine learning algorithm, data is analyzed based on existing training data to determine vehicle height adjustments suitable for new driving situations. KNN contributes to determining the optimal vehicle height difference value by comparing the input data with existing data similar to it.

[0166] (8) Extract inference results

[0167] The optimal suspension height is inferred based on data calculated by the overall height prediction unit and the machine learning model. This provides final data for vehicle height adjustment and reflects the vehicle's status in real time.

[0168] (9) Prediction-based risk extraction

[0169] The risk analysis unit uses inferred data to evaluate the risk of damage to the vehicle's undercarriage based on the current driving path and ground conditions. If the risk is high, it sends a signal to adjust the suspension height by a larger amount.

[0170] (10) Determination of total height control factors based on risk level

[0171] The suspension control unit determines the control factors required for vehicle height adjustment in real time based on data transmitted from the risk analysis unit. At this stage, an algorithm is applied to perform optimal height adjustment by comprehensively considering vehicle speed, ground conditions, etc.

[0172] (11) All height control

[0173] Finally, the suspension control unit adjusts the suspension height in real time based on the calculated overall height data. This process prevents damage to the vehicle's undercarriage, maintains driving stability, and improves ride comfort.

[0175] Figure 5 illustrates the suspension control process through risk analysis and KNN classification of a vehicle height adjustment system, showing the process in which a data learning unit and a risk analysis unit cooperate to perform vehicle height adjustment based on ground conditions and obstacle data in front of the vehicle.

[0176] The data learning unit utilizes the Kin Nearest Neighbors (KNN) machine learning algorithm to analyze risk based on existing data similar to the input data and derives optimal data for controlling the suspension height.

[0177] When an obstacle, such as a speed bump located in front of the vehicle, is detected, the data collection unit collects ground height and shape information. The data learning unit converts this information into vector features and applies the KNN algorithm. The KNN algorithm derives the optimal risk level by comparing the similarity between the vector features of the input data and the learned data. In Figure 5, each point (A, B, C, etc.) represents a learned dataset, and if the current vector features are close to B or F, the risk level (B or F) is determined.

[0178] The determined risk level is transmitted to the risk analysis unit, which comprehensively evaluates the ground conditions and the characteristics of obstacles. Through this, the vehicle height adjustment system controls the suspension height in real time to suit the current driving conditions, thereby preventing damage to the vehicle's undercarriage and maintaining driving safety.

[0180] FIG. 6 illustrates a risk-based suspension height optimization process according to an embodiment of the present invention, showing a process of optimizing the suspension height using a risk analysis unit and a suspension control unit. This process focuses on preventing damage to the vehicle undercarriage and maintaining driving stability through risk assessment and path optimization.

[0181] The risk analysis unit evaluates the risk based on obstacles or ground conditions in front of the vehicle and sets a correction weight corresponding to each risk level. In FIG. 6, risk levels (A, B, C, etc.) and corresponding correction weights are presented in a table format. For example, risk level A has a correction weight of 1 and risk level F has a correction weight of 10, and a larger weight is applied as the risk level increases.

[0182] Based on data received from the risk analysis unit, the suspension control unit determines the optimal suspension height by comparing the time required (fx) along the existing optimal path with a certain allowable margin (constant). If the calculated value exceeds the allowable margin, it instructs the suspension to be reset to the new optimal height (True). Conversely, if the calculated value is below the allowable margin, it maintains the existing path and height (False). This process ensures efficient ride height adjustment while maintaining the vehicle's driving stability.

[0184] In addition, the method according to an embodiment of the present invention may be implemented in the form of program instructions that can be executed through various computer means and may be recorded on a computer-readable medium.

[0185] The above computer-readable medium may include program instructions, data files, data structures, etc., either individually or in combination. The program instructions recorded on the computer-readable medium may be specially designed and configured for embodiments of the present invention, or they may be known and available to a person skilled in the art of computer software. The computer-readable recording medium may include a hardware device configured to store and execute program instructions. For example, the computer-readable recording medium may be magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; ROM; RAM; flash memory, etc. The program instructions may include not only machine code, such as that generated by a compiler, but also high-level language code that can be executed by a computer through an interpreter, etc.

[0186] Although embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements by those skilled in the art using the basic concept of the present invention as defined in the following claims also fall within the scope of the present invention.

Claims

Claim 1 A vehicle height adjustment system comprising: a ground sensing unit that detects the ground condition in front of the vehicle; a height prediction unit that predicts the vehicle height based on the detected ground data; and a height adjustment unit that adjusts the suspension height based on the predicted height data. Claim 2 In claim 1, the ground sensing unit is a vehicle height adjustment system configured to detect ground height, shape, and distance data in front of the vehicle. Claim 3 In paragraph 2, the ground sensing unit is configured to preprocess the detected data and convert it into an analyzable data format, thereby forming a vehicle height control system. Claim 4 In claim 1, the vehicle height control system configured such that the height prediction unit predicts changes in vehicle height according to ground conditions using a machine learning algorithm. Claim 5 In paragraph 4, the machine learning algorithm comprises a nearest neighbor (KNN) algorithm, a vehicle height control system. Claim 6 In claim 1, the above-described height adjustment unit is a vehicle height adjustment system configured to optimize suspension height using an algorithm. Claim 7 In claim 6, the above algorithm is a vehicle height adjustment system configured to apply distance weights for calculating path costs. Claim 8 A method for adjusting vehicle height, characterized by comprising: a step of detecting a ground condition in front of a vehicle; a step of preprocessing the detected ground data and converting it into an analyzable data format; a step of predicting the vehicle height based on the converted data; and a step of adjusting the suspension height based on the predicted data. Claim 9 In claim 8, the step of predicting the overall height comprises: a method for adjusting vehicle overall height that includes a process of predicting changes in vehicle overall height according to ground conditions using a machine learning algorithm. Claim 10 In claim 9, the machine learning algorithm comprises a vehicle height adjustment method including a nearest neighbor (KNN) algorithm. Claim 11 In claim 8, the step of adjusting the suspension height comprises a process of optimizing the suspension height using an algorithm, a method for adjusting the vehicle's overall height. Claim 12 In claim 11, the above algorithm is a vehicle height adjustment method that includes the process of applying distance weights for calculating path costs. Claim 13 A vehicle height adjustment system comprising: a data collection unit that collects ground condition and obstacle data through a camera positioned at the front of the vehicle; a risk analysis unit that analyzes data input from the data collection unit to evaluate the risk level of obstacles; a suspension control unit that adjusts the vehicle's suspension height in real time based on the data evaluated by the risk analysis unit; and a data learning unit that learns the data collected from the data collection unit and the risk analysis unit using a machine learning algorithm and generates a learning model for optimal height adjustment in various driving situations. Claim 14 In paragraph 13, the data collection unit is a vehicle height adjustment system configured to detect ground height, shape, and distance data in front of the vehicle. Claim 15 In Clause 13, the above-mentioned risk analysis unit is a vehicle height adjustment system configured to evaluate the possibility of vehicle undercarriage damage based on ground condition and obstacle data. Claim 16 In Clause 13, the suspension control unit is a vehicle height adjustment system configured to adjust the suspension height in real time according to the risk evaluated by the risk analysis unit. Claim 17 In claim 16, the suspension control unit is a vehicle height adjustment system configured to optimize the suspension height using an algorithm. Claim 18 In claim 17, the above algorithm is a vehicle height adjustment system configured to apply distance weights for calculating path costs. Claim 19 In claim 13, the data learning unit is configured to learn data collected through a machine learning algorithm and update the learning data so as to adjust the vehicle height according to the type of obstacle, thereby forming a vehicle height adjustment system. Claim 20 In claim 19, the machine learning algorithm comprises a vehicle height adjustment system including a nearest neighbor (KNN) algorithm.

Images