Vehicle mobility assessment method and system for off-road maneuverability

Abstract

This invention provides a vehicle maneuverability assessment method and system for off-road maneuvers, comprising: Step S1: identifying obstacles in an off-road environment and interpreting off-road geomechanical parameters; Step S2: inverting off-road geomechanical parameters for a coupled vehicle; Step S3: modeling vehicle dynamics based on vehicle-ground interaction; and Step S4: assessing vehicle maneuverability in an off-road environment using multivariate analysis and predicting maneuverability speed based on vehicle performance parameters and a dynamic model. This invention provides a more accurate assessment of vehicle maneuverability in off-road environments, comprehensively considering the vehicle's geometric and dynamic properties, such as turning radius, height, width, and weight.

Description

Technical Field

[0001] The present invention relates to information system software technology, and in particular to a vehicle maneuverability assessment method and system for off-road maneuvers. Background Art

[0002] Ground vehicle mobility primarily refers to the vehicle's ability to travel quickly under various road, ground, and terrain conditions it may encounter. Domestic research on off-road vehicle mobility has mostly focused on military vehicles. Although there are quantitative descriptions of vehicle trafficability assessments, these quantitative indicators are obtained through empirical values ​​or vehicle performance tests. If new vehicles or new terrain and geological conditions are encountered, it is difficult to extrapolate and obtain a quantitative vehicle trafficability assessment because no dynamic model has been constructed. This is mainly reflected in the following aspects:

[0003] (1) Mobility assessment relies mainly on driver experience, and the ability to quantitatively assist decision-making is insufficient. For example, the ability to quantitatively calculate the optimal safe speed, vehicle stability results, and rutting depth under each terrain and geological condition is insufficient; the factors considered are limited, such as only considering terrain factors, without considering weather, soil quality, natural disasters, and emergencies;

[0004] (2) The focus of mobility assessment is on roads, and the algorithms and models are not very practical. They are mainly based on road network data and simple shortest path analysis, which is far from the needs of off-road navigation. Off-road maneuverability and complex off-road environments are not considered enough, such as rain, snow, ice, strong winds, fog, and the resulting soil moisture and slipperiness.

[0005] (3) The results of land mobility assessments are not accurate, and modeling and prediction capabilities need to be greatly enhanced. For example, the modeling of vehicle dynamics, vehicle-soil interaction mechanics, and the estimation of maneuvering speed, fuel consumption, and travel time need to be further improved.

[0006] Zhang De, Zhang Yuepeng, Huang Limin, Yuan Bing. Digital off-road maneuverability analysis [J]. Surveying and Mapping Science and Engineering, 2017, 0(2): 44-4861. The literature states that off-road maneuverability analysis is of great significance for identifying maneuvering corridors, planning routes, and demarcating regional scopes, and is an important research topic in terrain analysis. This paper first discusses the current status of off-road maneuverability analysis at home and abroad, proposes a digital off-road maneuverability analysis process, and elaborates on its key technical methods, namely single-element layer maneuverability analysis and image morphological processing: single-element layer maneuverability analysis includes landform layer, water and land layer, residential area and ancillary facilities layer, and vegetation layer; image morphological processing includes dilation, erosion, filling, and attribute feature analysis. The method proposed in this paper is then experimentally verified.

[0007] However, early mobility assessment methods relied on empirical data accumulated through long-term testing. These data were used to derive empirical formulas to determine a vehicle's maneuverability within a specific area. These theoretical methods, based on empirical models, had some application value but were unable to fully explain the interaction between the vehicle and the ground, resulting in significant limitations.

[0008] Li Kunwei, You Xiong, Zhang Xin, Tang Fen. Soil off-road accessibility assessment based on multi-source data [J]. Journal of Surveying and Mapping Science and Technology, 2018, 35(02): 206-210. The literature states: In response to the shortcomings of existing soil accessibility assessment methods, a quantitative analysis of soil accessibility is conducted based on multi-source environmental data using the cone index as an indicator. High-precision USCS (Unified Soil Classification System) type soil data and rainfall data are the basis for quantitative analysis of soil accessibility. First, the random forest method is used to convert the existing soil data into USCS type; then, the meteorological station rainfall data and satellite rainfall data are fused to construct high-precision rainfall data; finally, the cone index of the soil is calculated using the cone index model, and the soil accessibility is evaluated by comparing the vehicle cone index and the soil cone index.

[0009] However, the paper proposes a soil off-road passability assessment method based on multi-source data. While the method comprehensively assesses the impact of off-road conditions on passability, it fails to provide differentiated evaluations for vehicles with varying mobility capabilities, and only offers a general, overarching analysis. Furthermore, the paper only analyzes vehicle passability and lacks analysis of other mobility capabilities, such as vehicle speed, fuel consumption, and travel time estimates in off-road environments.

[0010] Han Yu, Meng Guangwei, Huang Chaosheng, Men Yuzhuo. Research on the maneuverability of off-road vehicles[J]. Vibration and Shock, 2015, 34(02): 96-100. DOI: 10.13465 / j.cnki.jvs.2015.02.017. Literature records: In order to study the maneuverability of off-road vehicles on uneven roads, some characteristic parameters of the vehicle were identified through frequency deviation test, the suspension displacement of the test vehicle was measured, the road surface roughness of the driving section was calculated, the time series signal of the driver's seat Z acceleration on the uneven test section was monitored, the human body suction power was calculated according to the human body fatigue characteristics, and a "human body-body-wheel" three-degree-of-freedom vibration model was established; the time series signal of the uneven road section was generated by Fourier inverse transform as simulation input, and the time series signal of the Z acceleration of the test vehicle driver's seat on the uneven road section was simulated using Simulink; the vehicle speed corresponding to the human body suction power of 6W was proposed as the evaluation method for the maneuverability of off-road vehicles on uneven roads. However, this document only considers the impact of vehicle smoothness on vehicle maneuverability, and the evaluation factors are too simple.

[0011] Patent document CN113984062A (application number: CN202111250187.7) discloses a ground vehicle path planning method based on mobility assessment, including obtaining multi-source data and using geostatistical methods to reconstruct the terrain and environment in three dimensions to obtain a three-dimensional point cloud terrain model. The multi-source data includes remote sensing elevation terrain data, land use data, soil type distribution data, and vehicle data; based on the passability analysis and mobility cost quantification of terrain factors and ground mechanical effects, the vehicle mobility of the passable area is evaluated; and the cost function of the improved A-Star algorithm is used for path planning. However, this invention relies solely on road information and does not comprehensively consider the terrain, weather, soil quality, vegetation, real-time weather forecast information, bridge bearing capacity, tunnels and other multi-source heterogeneous and complex off-road environmental information within the off-road area. Summary of the Invention

[0012] In view of the defects in the prior art, the purpose of the present invention is to provide a vehicle maneuverability assessment method and system for off-road maneuvers.

[0013] A vehicle maneuverability assessment method for off-road maneuvers provided by the present invention includes:

[0014] Step S1: Identify obstacles in the off-road environment and interpret the off-road environment ground mechanics parameters;

[0015] Step S2: Inversion of the off-road environment ground mechanical parameters of the coupled vehicle;

[0016] Step S3: Modeling the vehicle dynamics based on the interaction between the vehicle and the ground;

[0017] Step S4: Evaluate the vehicle's passability in an off-road environment based on multivariate analysis and perform maneuverability speed prediction based on vehicle performance parameters and dynamics models.

[0018] Preferably, in step S1:

[0019] Identify obstacles in off-road environments:

[0020] Build an intelligent obstacle recognition model based on scene scale and DCNN transfer learning. Utilize public datasets and DCNN transfer learning strategies to use scenes as primitives to represent the overall characteristics of obstacles and construct complex semantic patterns.

[0021] Interpretation of ground mechanics parameters in off-road environments:

[0022] Construct a rock and soil classification system for ground mechanical properties, survey and identify ground rock and soil characteristic types, and construct ground soil mechanical property parameters for off-road environments based on the soil moisture inversion method based on remote sensing data.

[0023] Preferably, in step S2:

[0024] Establish a knowledge base of the corresponding mechanical properties of known soil types, obtain a coupled quantitative calculation method that meets the ground mechanical property parameters of different passing vehicles and standard classifications, and realize the inversion of ground mechanical property parameters in unknown areas and the evaluation of vehicle traffic performance.

[0025] Preferably, in step S3:

[0026] The vehicle-ground interaction model is determined by the type of vehicle and ground. Different types of ground have different characterization parameters. The type of vehicle is determined by the type of its running mechanism, and the type of ground exists in many different ways due to its composition.

[0027] Based on the interaction mechanism between different walking mechanisms and the ground, a multi-degree-of-freedom mechanical model of the vehicle in a three-dimensional unstructured environment is established, providing a model basis for the evaluation of vehicle passability and maneuverability.

[0028] Preferably, in step S4:

[0029] Evaluating vehicle trafficability in off-road environments based on multivariate analysis:

[0030] Vehicle passability refers to the ability of a vehicle to pass through a specified area. Passability assessment research must first parametrically characterize the ground surface. This involves integrating road surface information, terrain information, landform information, and climate information. Feature extraction and selection are performed based on a combination of subjective and objective methods to obtain feature types with a correlation with vehicle passability that exceeds a preset value. This allows the construction of a comprehensive ground characterization parameter model encompassing both geometric and geological characteristics.

[0031] In terms of vehicle characteristics, the vehicle's own characteristics are deconstructed and analyzed in combination with theoretical design or experimental experience. The theoretical design includes structural dimensions, dynamic characteristics, and performance indicators. The factors affecting passability are studied in combination with a comprehensive ground characterization parameterized model. The interaction mechanism between vehicle parameters, ground characterization, and passability is revealed, a quantitative analysis relationship is established, and an evaluation model for multi-dimensional judgment is formed.

[0032] Predicting maneuverability speed in off-road environments based on vehicle performance parameters and dynamic models:

[0033] Maneuverability assessment is an assessment of vehicle operating speed. On the basis of passability assessment, for different types of vehicles and surfaces, based on the vehicle power transmission characteristics and the vehicle-surface coupling dynamics model, a maneuvering speed prediction method based on the power flow model is studied; for vehicle operation safety, a driving stability classification study based on deep learning is conducted in combination with vehicle test data, the ground comprehensive characterization model and the vehicle maneuvering speed; for the judgment of the rapid maneuvering area, a study of the rapid maneuvering rules is conducted in combination with vehicle stability and speed, and a rapid maneuvering area evaluation model is formed, thereby realizing the judgment of the rapid maneuvering area.

[0034] A vehicle maneuverability assessment system for off-road maneuvers provided by the present invention includes:

[0035] Module M1: Identify obstacles in off-road environments and interpret off-road environment ground mechanics parameters;

[0036] Module M2: Inversion of ground mechanics parameters in off-road environments with coupled vehicles;

[0037] Module M3: Modeling of vehicle dynamics based on vehicle-ground interaction;

[0038] Module M4: Evaluate vehicle trafficability in off-road environments based on multivariate analysis and predict maneuverability speed based on vehicle performance parameters and dynamic models.

[0039] Preferably, in the module M1:

[0040] Identify obstacles in off-road environments:

[0041] Build an intelligent obstacle recognition model based on scene scale and DCNN transfer learning. Utilize public datasets and DCNN transfer learning strategies to use scenes as primitives to represent the overall characteristics of obstacles and construct complex semantic patterns.

[0042] Interpretation of ground mechanics parameters in off-road environments:

[0043] Construct a rock and soil classification system for ground mechanical properties, survey and identify ground rock and soil characteristic types, and construct ground soil mechanical property parameters for off-road environments based on the soil moisture inversion method based on remote sensing data.

[0044] Preferably, in the module M2:

[0045] Establish a knowledge base of the corresponding mechanical properties of known soil types, obtain a coupled quantitative calculation method that meets the ground mechanical property parameters of different passing vehicles and standard classifications, and realize the inversion of ground mechanical property parameters in unknown areas and the evaluation of vehicle traffic performance.

[0046] Preferably, in the module M3:

[0047] The vehicle-ground interaction model is determined by the type of vehicle and ground. Different types of ground have different characterization parameters. The type of vehicle is determined by the type of its running mechanism, and the type of ground exists in many different ways due to its composition.

[0048] Based on the interaction mechanism between different walking mechanisms and the ground, a multi-degree-of-freedom mechanical model of the vehicle in a three-dimensional unstructured environment is established, providing a model basis for the evaluation of vehicle passability and maneuverability.

[0049] Preferably, in the module M4:

[0050] Evaluating vehicle trafficability in off-road environments based on multivariate analysis:

[0051] Vehicle passability refers to the ability of a vehicle to pass through a specified area. Passability assessment research must first parametrically characterize the ground surface. This involves integrating road surface information, terrain information, landform information, and climate information. Feature extraction and selection are performed based on a combination of subjective and objective methods to obtain feature types with a correlation with vehicle passability that exceeds a preset value. This allows the construction of a comprehensive ground characterization parameter model encompassing both geometric and geological characteristics.

[0052] In terms of vehicle characteristics, the vehicle's own characteristics are deconstructed and analyzed in combination with theoretical design or experimental experience. The theoretical design includes structural dimensions, dynamic characteristics, and performance indicators. The factors affecting passability are studied in combination with a comprehensive ground characterization parameterized model. The interaction mechanism between vehicle parameters, ground characterization, and passability is revealed, a quantitative analysis relationship is established, and an evaluation model for multi-dimensional judgment is formed.

[0053] Predicting maneuverability speed in off-road environments based on vehicle performance parameters and dynamic models:

[0054] Maneuverability assessment is an assessment of vehicle operating speed. On the basis of passability assessment, for different types of vehicles and surfaces, based on the vehicle power transmission characteristics and the vehicle-surface coupling dynamics model, a maneuvering speed prediction method based on the power flow model is studied; for vehicle operation safety, a driving stability classification study based on deep learning is conducted in combination with vehicle test data, the ground comprehensive characterization model and the vehicle maneuvering speed; for the judgment of the rapid maneuvering area, a study of the rapid maneuvering rules is conducted in combination with vehicle stability and speed, and a rapid maneuvering area evaluation model is formed, thereby realizing the judgment of the rapid maneuvering area.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. This invention provides a more comprehensive assessment of vehicle maneuverability in off-road environments. It no longer relies solely on road information. Instead, it comprehensively considers multiple sources of heterogeneous and complex off-road environmental information, including terrain, weather, soil quality, vegetation, real-time weather forecasts, bridge load-bearing capacity, and tunnels within the off-road area, to intelligently identify obstacles in the off-road environment and interpret ground mechanics parameters.

[0057] 2. This invention provides a more accurate assessment of vehicle maneuverability in off-road environments, taking into account the vehicle's geometric and dynamic properties, such as turning radius, height, width, and weight.

[0058] 3. The present invention provides more comprehensive and practical results for evaluating vehicle maneuverability in off-road environments. By conducting a multivariate passability analysis of the interaction between the vehicle and the ground, the vehicle's passability in a real off-road environment can be evaluated based on real-time off-road environmental information and the type of motor vehicle. Specifically, the vehicle's passability is evaluated, including whether it can pass, the optimal safe speed, fuel consumption, rutting depth, and other content. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0060] Figure 1 Provide a technical roadmap for intelligent identification methods of geotechnical masses;

[0061] Figure 2 Identify network structures for rock properties taking into account local and global information;

[0062] Figure 3 It is a ground mechanical characteristics inversion model diagram that combines ground classification and parameter identification;

[0063] Figure 4 Schematic diagram for the dynamic modeling of the walking mechanism and the entire vehicle;

[0064] Figure 5 Evaluate the technology roadmap for passability;

[0065] Figure 6 Predicting technology roadmaps for mobility speed. DETAILED DESCRIPTION

[0066] The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the scope of the present invention. These all fall within the scope of protection of the present invention.

[0067] Example 1:

[0068] In view of the defects in the prior art, the technical problems to be solved by the present invention are embodied in the following points:

[0069] 1) The consideration of off-road environmental factors in mobility assessment is relatively limited and not comprehensive;

[0070] 2) The vehicle's dynamic properties, such as driving force, climbing ability, obstacle crossing ability, and braking ability, are not considered when evaluating mobility;

[0071] 3) Mobility assessment results are incomplete and inaccurate, and modeling and prediction capabilities need to be significantly strengthened. For example, the ability to estimate vehicle maneuvering speed, fuel consumption, and travel time in off-road environments needs further improvement.

[0072] According to the present invention, a vehicle maneuverability assessment method for off-road maneuvers is provided. Figures 1-6 Shown, including:

[0073] Step S1: Identify obstacles in the off-road environment and interpret the off-road environment ground mechanics parameters;

[0074] Specifically, in step S1:

[0075] Identify obstacles in off-road environments:

[0076] Build an intelligent obstacle recognition model based on scene scale and DCNN transfer learning. Utilize public datasets and DCNN transfer learning strategies to use scenes as primitives to represent the overall characteristics of obstacles and construct complex semantic patterns.

[0077] Interpretation of ground mechanics parameters in off-road environments:

[0078] Construct a rock and soil classification system for ground mechanical properties, survey and identify ground rock and soil characteristic types, and construct ground soil mechanical property parameters for off-road environments based on the soil moisture inversion method based on remote sensing data.

[0079] Step S2: Inversion of the off-road environment ground mechanical parameters of the coupled vehicle;

[0080] Specifically, in step S2:

[0081] Establish a knowledge base of the corresponding mechanical properties of known soil types, obtain a coupled quantitative calculation method that meets the ground mechanical property parameters of different passing vehicles and standard classifications, and realize the inversion of ground mechanical property parameters in unknown areas and the evaluation of vehicle traffic performance.

[0082] Step S3: Modeling the vehicle dynamics based on the interaction between the vehicle and the ground;

[0083] Specifically, in step S3:

[0084] The vehicle-ground interaction model is determined by the type of vehicle and ground. Different types of ground have different characterization parameters. The type of vehicle is determined by the type of its running mechanism, and the type of ground exists in many different ways due to its composition.

[0085] Based on the interaction mechanism between different walking mechanisms and the ground, a multi-degree-of-freedom mechanical model of the vehicle in a three-dimensional unstructured environment is established, providing a model basis for the evaluation of vehicle passability and maneuverability.

[0086] Step S4: Evaluate the vehicle's passability in an off-road environment based on multivariate analysis and perform maneuverability speed prediction based on vehicle performance parameters and dynamics models.

[0087] Specifically, in step S4:

[0088] Evaluating vehicle trafficability in off-road environments based on multivariate analysis:

[0089] Vehicle passability refers to the ability of a vehicle to pass through a specified area. Passability assessment research must first parametrically characterize the ground surface. This involves integrating road surface information, terrain information, landform information, and climate information. Feature extraction and selection are performed based on a combination of subjective and objective methods to obtain feature types with a correlation with vehicle passability that exceeds a preset value. This allows the construction of a comprehensive ground characterization parameter model encompassing both geometric and geological characteristics.

[0090] In terms of vehicle characteristics, the vehicle's own characteristics are deconstructed and analyzed in combination with theoretical design or experimental experience. The theoretical design includes structural dimensions, dynamic characteristics, and performance indicators. The factors affecting passability are studied in combination with a comprehensive ground characterization parameterized model. The interaction mechanism between vehicle parameters, ground characterization, and passability is revealed, a quantitative analysis relationship is established, and an evaluation model for multi-dimensional judgment is formed.

[0091] Predicting maneuverability speed in off-road environments based on vehicle performance parameters and dynamic models:

[0092] Maneuverability assessment is an assessment of vehicle operating speed. On the basis of passability assessment, for different types of vehicles and surfaces, based on the vehicle power transmission characteristics and the vehicle-surface coupling dynamics model, a maneuvering speed prediction method based on the power flow model is studied; for vehicle operation safety, a driving stability classification study based on deep learning is conducted in combination with vehicle test data, the ground comprehensive characterization model and the vehicle maneuvering speed; for the judgment of the rapid maneuvering area, a study of the rapid maneuvering rules is conducted in combination with vehicle stability and speed, and a rapid maneuvering area evaluation model is formed, thereby realizing the judgment of the rapid maneuvering area.

[0093] Example 2:

[0094] Example 2 is a preferred example of Example 1 and is used to illustrate the present invention in more detail.

[0095] The present invention also provides a XX system, which can be implemented by executing the process steps of the XX method. That is, those skilled in the art can understand the XX method as a preferred embodiment of the XX system.

[0096] A vehicle maneuverability assessment system for off-road maneuvers provided by the present invention includes:

[0097] Module M1: Identify obstacles in off-road environments and interpret off-road environment ground mechanics parameters;

[0098] Specifically, in the module M1:

[0099] Identify obstacles in off-road environments:

[0100] Build an intelligent obstacle recognition model based on scene scale and DCNN transfer learning. Utilize public datasets and DCNN transfer learning strategies to use scenes as primitives to represent the overall characteristics of obstacles and construct complex semantic patterns.

[0101] Interpretation of ground mechanics parameters in off-road environments:

[0102] Construct a rock and soil classification system for ground mechanical properties, survey and identify ground rock and soil characteristic types, and construct ground soil mechanical property parameters for off-road environments based on the soil moisture inversion method based on remote sensing data.

[0103] Module M2: Inversion of ground mechanics parameters in off-road environments with coupled vehicles;

[0104] Specifically, in the module M2:

[0105] Establish a knowledge base of the corresponding mechanical properties of known soil types, obtain a coupled quantitative calculation method that meets the ground mechanical property parameters of different passing vehicles and standard classifications, and realize the inversion of ground mechanical property parameters in unknown areas and the evaluation of vehicle traffic performance.

[0106] Module M3: Modeling of vehicle dynamics based on vehicle-ground interaction;

[0107] Specifically, in the module M3:

[0108] The vehicle-ground interaction model is determined by the type of vehicle and ground. Different types of ground have different characterization parameters. The type of vehicle is determined by the type of its running mechanism, and the type of ground exists in many different ways due to its composition.

[0109] Based on the interaction mechanism between different walking mechanisms and the ground, a multi-degree-of-freedom mechanical model of the vehicle in a three-dimensional unstructured environment is established, providing a model basis for the evaluation of vehicle passability and maneuverability.

[0110] Module M4: Evaluate vehicle trafficability in off-road environments based on multivariate analysis and predict maneuverability speed based on vehicle performance parameters and dynamic models.

[0111] Specifically, in the module M4:

[0112] Evaluating vehicle trafficability in off-road environments based on multivariate analysis:

[0113] Vehicle passability refers to the ability of a vehicle to pass through a specified area. Passability assessment research must first parametrically characterize the ground surface. This involves integrating road surface information, terrain information, landform information, and climate information. Feature extraction and selection are performed based on a combination of subjective and objective methods to obtain feature types with a correlation with vehicle passability that exceeds a preset value. This allows the construction of a comprehensive ground characterization parameter model encompassing both geometric and geological characteristics.

[0114] In terms of vehicle characteristics, the vehicle's own characteristics are deconstructed and analyzed in combination with theoretical design or experimental experience. The theoretical design includes structural dimensions, dynamic characteristics, and performance indicators. The factors affecting passability are studied in combination with a comprehensive ground characterization parameterized model. The interaction mechanism between vehicle parameters, ground characterization, and passability is revealed, a quantitative analysis relationship is established, and an evaluation model for multi-dimensional judgment is formed.

[0115] Predicting maneuverability speed in off-road environments based on vehicle performance parameters and dynamic models:

[0116] Maneuverability assessment is an assessment of vehicle operating speed. On the basis of passability assessment, for different types of vehicles and surfaces, based on the vehicle power transmission characteristics and the vehicle-surface coupling dynamics model, a maneuvering speed prediction method based on the power flow model is studied; for vehicle operation safety, a driving stability classification study based on deep learning is conducted in combination with vehicle test data, the ground comprehensive characterization model and the vehicle maneuvering speed; for the judgment of the rapid maneuvering area, a study of the rapid maneuvering rules is conducted in combination with vehicle stability and speed, and a rapid maneuvering area evaluation model is formed, thereby realizing the judgment of the rapid maneuvering area.

[0117] Example 3:

[0118] Example 3 is a preferred example of Example 1 and is used to illustrate the present invention in more detail.

[0119] This method comprises the following steps:

[0120] Step 1: Intelligent identification of obstacles in off-road environments

[0121] To address the need for rapid obstacle recognition in off-road environments, we are developing intelligent obstacle recognition in off-road environments based on scene scale and DCNN transfer learning. The high semantic level and clear physical boundaries make intelligent interpretation difficult, and the small number of samples required requires an appropriate scale and overall feature representation. Therefore, to address the challenges of low intelligence and accuracy in obstacle recognition, we are developing an intelligent obstacle recognition model based on scene scale and DCNN transfer learning. Leveraging publicly available datasets and DCNN transfer learning strategies, we use scenes as primitives to characterize the overall features of obstacles and construct complex semantic patterns.

[0122] Step 2: Interpretation of off-road environment ground mechanics parameters

[0123] Interpreting geomechanical parameters in off-road environments provides the data foundation for coupled vehicle-geomechanical analysis. Geomechanical properties in off-road environments are significantly influenced by soil type and soil moisture. To interpret geomechanical parameters in off-road environments, soil classification and identification are necessary, along with soil moisture information, to construct geomechanical parameters. This research includes developing a geomechanical geotechnical classification system, rapidly surveying and identifying geotechnical characteristics, inverting soil moisture using remote sensing data, and constructing geomechanical parameters for off-road environments.

[0124] Step 3: Inversion of off-road environment ground mechanics parameters of coupled vehicle

[0125] In order to quantitatively describe the relationship between ground rock and soil types and their corresponding mechanical properties, as well as the mechanical characteristics corresponding to different vehicle types, the mechanical characteristics of soil types that affect vehicle traffic are studied, a knowledge base of the corresponding mechanical properties of known soil types is established, and a coupled quantitative calculation method that meets the requirements of different passing vehicles and standard classification of ground mechanical properties is obtained, thereby realizing the inversion of ground mechanical properties parameters in unknown areas and the evaluation of vehicle traffic performance.

[0126] Step 4: Full vehicle dynamics modeling based on vehicle-ground interaction

[0127] The vehicle-terrain interaction model is primarily determined by the type of vehicle and terrain. Different terrain types are characterized by distinct parameters. The type of vehicle is primarily determined by the type of its running gear, while terrain types vary depending on their composition. Therefore, studying the interaction mechanisms between vehicles and different terrain types can effectively reflect these interactions and derive the corresponding mechanical parameters for both vehicles and terrain, providing guidance for the inversion of terrain mechanical parameters. Based on the interaction mechanisms between different running gears and terrains, a multi-degree-of-freedom mechanical model of a vehicle in a three-dimensional unstructured environment is established, providing a model basis for evaluating vehicle passability and maneuverability.

[0128] Step 5: Vehicle passability evaluation in off-road environment based on multivariate analysis

[0129] Vehicle passability primarily refers to a vehicle's ability to pass through a designated area and is a GO / NOGO decision. Passability assessment research must first parametrically characterize the ground surface. This involves integrating road surface information, terrain information, landform information, and climate information. Feature extraction and selection are then performed based on a combination of subjective and objective methods to obtain features highly correlated with vehicle passability. This allows for the construction of a comprehensive ground characterization parameter model encompassing both geometric and geological characteristics. Regarding the vehicle's inherent characteristics, theoretical design or experimental experience, such as structural dimensions, dynamic characteristics, and performance indicators, is combined to deconstruct and analyze the vehicle's inherent characteristics. This is then combined with the comprehensive ground characterization parameter model to study the factors and indicators that influence passability. This reveals the interaction between vehicle parameters, ground characterization, and passability, establishes a quantitative analysis relationship, and forms an evaluation model for multi-dimensional judgment.

[0130] Step 6: Prediction of maneuverability speed based on vehicle performance parameters and dynamic model in off-road environment

[0131] Maneuverability assessment primarily focuses on the evaluation of vehicle operating speed. Building on passability assessment, research is being conducted on maneuvering speed prediction methods based on power flow models, tailored to different vehicle and surface types, and informed by vehicle powertrain characteristics and vehicle-surface coupling dynamics models. For vehicle operational safety, deep learning-based driving stability classification research is being conducted, combining vehicle test data, comprehensive surface characterization models, and vehicle maneuvering speed. Research on the identification of rapid maneuvering areas requires combining vehicle stability and speed, integrating these two factors to develop rules for rapid maneuverability and develop a rapid maneuverability area assessment model, thereby enabling the identification of rapid maneuverability areas.

[0132] Example 4:

[0133] Example 4 is a preferred example of Example 1 and is used to illustrate the present invention in more detail.

[0134] The vehicle-terrain mechanics model is the theoretical basis for studying the interaction between vehicles and the ground. Tracked and wheeled walking mechanisms correspond to track-ground and wheeled-ground mechanics respectively. Based on Bekker theory, the interaction models of different walking mechanisms and soil are established (e.g. Figure 4 As shown). The interaction model between the walking mechanism and the soil mainly includes the pressure-settlement characterization model (pressure-bearing characteristics) and the shear stress and shear displacement characterization model (shear characteristics). Both of the above characterizations have a strong correlation with the ground type. For the ground composed of general average soil, its main parameters are: pressure-related parameters include k_c (cohesion modulus), k_v (friction modulus), n (deformation index); shear-related parameters include K (shear modulus), c (ground cohesion coefficient), φ (internal shear strength angle); soil-related parameters include moisture content, cone index, soil density, etc.

[0135] The pressure-sinking characterization model is used to determine the compaction resistance of the soil to a single traveling mechanism. The driving force of a vehicle on soft soil is generated by the shearing of the ground by the traveling mechanism. Therefore, based on the characterization of the shearing characteristics, the vehicle's driving characteristics can be determined. A three-dimensional vehicle dynamics model is established based on the interaction model between the traveling mechanism and the soil. This allows the pressure distribution between each sub-traveling mechanism of the vehicle and the ground to be determined, and the sinking of each traveling mechanism can be further determined. Rutting depth is the sinking amount. Using the state quantity of the vehicle's position as input, the bearing relationship between each traveling mechanism on the vehicle and the ground is determined, and the sinking amount is calculated based on the pressure-sinking characterization. By combining the pressure-bearing and shearing characteristics, and taking into account factors such as slope resistance and headwind resistance encountered during vehicle operation, the resistance encountered by the vehicle during driving is determined, providing a model foundation for passability assessment and maneuverability speed prediction.

[0136] Regarding the evaluation of regional passability (GO / NOGO), there are two main factors that restrict vehicle passage: obstacle factors and dynamic factors. Obstacle factors mainly refer to the existence of obstacles that meet certain conditions in the area, and vehicles cannot pass through such obstacles directly or by detouring; dynamic factors refer to the power required for vehicles to pass. In addition, vehicles also have problems such as insufficient energy storage, road damage caused by multiple passes, etc., which may cause them to be unable to pass. Therefore, the passability evaluation technology mainly starts from the vehicle, the environment and their interaction, and combines experience and knowledge to achieve multivariate analysis. Technical routes such as Figure 5 shown.

[0137] First, the system acquires information about the area's surface type, geomechanical parameters, obstacle types and characteristic parameters, altitude, climate, and slope. A comprehensive ground model is then constructed to create a parametric representation of the area. Then, based on the vehicle-ground coupled dynamics model, the system inputs vehicle parameters, geomechanical parameters, and slope information to produce the outputs of the driving force required to pass through the area, the depth of the resulting ruts, and the estimated energy consumption.

[0138] For obstacle factors, obstacle determination criteria are established based on empirical knowledge of the vehicle's obstacle-crossing capabilities (e.g., trenches, vertical walls, wading depths, and other experimental experience or theoretical design values). These criteria primarily assess the ability to traverse obstacles within the area. For dynamic factors, dynamic determination criteria are established based on vehicle power and transmission characteristics data. These criteria primarily assess the degree to which the vehicle's available power matches the power required to navigate the area to be entered, thereby determining dynamic performance. Furthermore, for other factors, determination criteria are established based on estimated energy consumption and rutting depth. These criteria primarily assess the energy balance required to navigate the area to be entered and the extent of terrain damage. Therefore, combining these three determination criteria, a multivariate passability determination model is developed. This model is divided into two inference processes: direct and indirect. Direct determination is used for areas with significant topography; obstacle determination criteria can directly determine the passability of the area. For areas with largely flat, non-deformed surfaces, passability can also be directly determined. Indirect judgment is mainly aimed at areas where direct judgment cannot be confirmed. It is necessary to obtain the road surface characterization parameter model of the area, combine it with the vehicle-ground coupling dynamics model, and mainly use dynamic judgment conditions to further judge the vehicle's passability.

[0139] Prediction of maneuverability speed based on vehicle performance parameters and dynamic model in off-road environment:

[0140] The assessment of maneuverability speed is mainly based on the assessment of passability. It can be predicted by empirical method or calculation method according to the terrain and geological characteristics of the passable area. Figure 6 shown.

[0141] For relatively flat, non-deformable or less deformed roads (i.e., directly judging the drivable area), and typical water terrain with obvious characteristics such as rivers and lakes, the vehicle's speed experience data (maximum off-road average speed, maximum speed, underwater maneuvering speed, etc.) can be directly referenced to realize the maneuverability speed assessment.

[0142] For deformed roads and areas with significant obstacles (such as slopes and mudflats), a dynamic model and the vehicle's powertrain must be combined to construct a power flow model, which is then used as a basis for speed estimation. Based on the engine's external characteristic curve or the motor's characteristic curve, the speed-torque relationship for different gear conditions is derived. Furthermore, the vehicle-ground coupled dynamic model to be evaluated can be used to determine the resistance that the vehicle must overcome, thereby determining the required torque and the speed range within which it is located. This allows the vehicle's permissible maneuvering speed range to be predicted, and furthermore, the vehicle's maximum permissible maneuvering speed in that area can be determined.

[0143] The evaluation of required speeds, specifically the optimal safe speed and the area where rapid maneuvering is permitted, is primarily determined by combining ground conditions, speed, and experience. Based on a comprehensive ground characterization model, empirical vehicle test data, and permitted maneuvering speeds, a deep neural network classification method is used to categorize and evaluate vehicle operational stability at different maneuvering speeds (smooth, bumpy, and very bumpy). The speed with the highest stability rating, or optimal safe speed, is determined. Furthermore, combining vehicle stability evaluation classifications and maneuvering speed ranges, rapid maneuvering rules are established, and areas that meet these rules are classified as areas where rapid maneuvering is permitted.

[0144] Those skilled in the art will appreciate that, in addition to implementing the system, device, and various modules provided by the present invention in purely computer-readable program code, it is entirely possible to implement the same program in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, and the like by logically programming the method steps. Therefore, the system, device, and various modules provided by the present invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; the modules for implementing various functions can also be considered both software programs for implementing the method and structures within the hardware component.

[0145] The above describes specific embodiments of the present invention. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art may make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. The embodiments of this application and the features in the embodiments may be combined with each other in any manner unless there is a conflict.

Claims

1. A vehicle maneuverability assessment method for off-road maneuvers, characterized in that: include: Step S1: Identify obstacles in the off-road environment and interpret the off-road environment ground mechanics parameters; Step S2: Inversion of the off-road environment ground mechanical parameters of the coupled vehicle; Step S3: Modeling the vehicle dynamics based on the interaction between the vehicle and the ground; Step S4: evaluating the vehicle's passability in an off-road environment based on multivariate analysis and performing maneuverability speed prediction based on vehicle performance parameters and a dynamic model; In step S4: Evaluating vehicle trafficability in off-road environments based on multivariate analysis: Vehicle passability refers to the ability of a vehicle to pass through a specified area. Passability assessment research must first parametrically characterize the ground surface. This involves integrating road surface information, terrain information, landform information, and climate information. Feature extraction and selection are performed based on a combination of subjective and objective methods to obtain feature types with a correlation with vehicle passability that exceeds a preset value. This allows the construction of a comprehensive ground characterization parameter model encompassing both geometric and geological characteristics. In terms of vehicle characteristics, the vehicle's own characteristics are deconstructed and analyzed in combination with theoretical design or experimental experience. The theoretical design includes structural dimensions, dynamic characteristics, and performance indicators. The factors affecting passability are studied in combination with a comprehensive ground characterization parameterized model. The interaction mechanism between vehicle parameters, ground characterization, and passability is revealed, a quantitative analysis relationship is established, and an evaluation model for multi-dimensional judgment is formed. Predicting maneuverability speed in off-road environments based on vehicle performance parameters and dynamic models: Maneuverability assessment is an evaluation of vehicle operating speed. Based on the passability assessment, for different types of vehicles and surfaces, based on the vehicle's power transmission characteristics and the vehicle-surface coupling dynamics model, we conduct research on maneuverability speed prediction methods based on power flow models. For vehicle operational safety, we conduct deep learning-based driving stability classification research by combining vehicle test data, comprehensive surface characterization models, and vehicle maneuverability speed. The study on the judgment of the area where rapid maneuvering is possible is carried out by combining the stability and speed of the vehicle to carry out the study of the rules for rapid maneuvering, forming an evaluation model for the area where rapid maneuvering is possible, thereby realizing the judgment of the area where rapid maneuvering is possible.

2. The vehicle maneuverability assessment method for off-road maneuvers according to claim 1, characterized in that: In step S1: Identify obstacles in off-road environments: Build an intelligent obstacle recognition model based on scene scale and DCNN transfer learning. Utilize public datasets and DCNN transfer learning strategies to use scenes as primitives to represent the overall characteristics of obstacles and construct complex semantic patterns. Interpretation of ground mechanics parameters in off-road environments: Construct a rock and soil classification system for ground mechanical properties, survey and identify ground rock and soil characteristic types, and construct ground soil mechanical property parameters for off-road environments based on the soil moisture inversion method based on remote sensing data.

3. The vehicle maneuverability assessment method for off-road maneuvers according to claim 1, characterized in that: In step S2: Establish a knowledge base of the corresponding mechanical properties of known soil types, obtain a coupled quantitative calculation method that meets the ground mechanical property parameters of different passing vehicles and standard classifications, and realize the inversion of ground mechanical property parameters in unknown areas and the evaluation of vehicle traffic performance.

4. The vehicle maneuverability assessment method for off-road maneuvers according to claim 1, characterized in that: In step S3: The vehicle-ground interaction model is determined by the type of vehicle and ground. Different types of ground have different characterization parameters. The type of vehicle is determined by the type of its running mechanism, and the type of ground exists in many different ways due to its composition. Based on the interaction mechanism between different walking mechanisms and the ground, a multi-degree-of-freedom mechanical model of the vehicle in a three-dimensional unstructured environment is established, providing a model basis for the evaluation of vehicle passability and maneuverability.

5. A vehicle maneuverability assessment system for off-road maneuvers, characterized in that: include: Module M1: Identify obstacles in off-road environments and interpret off-road environment ground mechanics parameters; Module M2: Inversion of ground mechanics parameters in off-road environments with coupled vehicles; Module M3: Modeling of vehicle dynamics based on vehicle-ground interaction; Module M4: Evaluate vehicle trafficability in off-road environments based on multivariate analysis and predict maneuverability speed based on vehicle performance parameters and dynamic models; In the module M4: Evaluating vehicle trafficability in off-road environments based on multivariate analysis: Vehicle passability refers to the ability of a vehicle to pass through a specified area. Passability assessment research must first parametrically characterize the ground surface. This involves integrating road surface information, terrain information, landform information, and climate information. Feature extraction and selection are performed based on a combination of subjective and objective methods to obtain feature types with a correlation with vehicle passability that exceeds a preset value. This allows the construction of a comprehensive ground characterization parameter model encompassing both geometric and geological characteristics. In terms of vehicle characteristics, the vehicle's own characteristics are deconstructed and analyzed in combination with theoretical design or experimental experience. The theoretical design includes structural dimensions, dynamic characteristics, and performance indicators. The factors affecting passability are studied in combination with a comprehensive ground characterization parameterized model. The interaction mechanism between vehicle parameters, ground characterization, and passability is revealed, a quantitative analysis relationship is established, and an evaluation model for multi-dimensional judgment is formed. Predicting maneuverability speed in off-road environments based on vehicle performance parameters and dynamic models: Maneuverability assessment is an evaluation of vehicle operating speed. Based on the passability assessment, for different types of vehicles and surfaces, based on the vehicle's power transmission characteristics and the vehicle-surface coupling dynamics model, we conduct research on maneuverability speed prediction methods based on power flow models. For vehicle operational safety, we conduct deep learning-based driving stability classification research by combining vehicle test data, comprehensive surface characterization models, and vehicle maneuverability speed. The study on the judgment of the area where rapid maneuvering is possible is carried out by combining the stability and speed of the vehicle to carry out the study of the rules for rapid maneuvering, forming an evaluation model for the area where rapid maneuvering is possible, thereby realizing the judgment of the area where rapid maneuvering is possible.

6. The vehicle maneuverability assessment system for off-road maneuvers according to claim 5, characterized in that: In the module M1: Identify obstacles in off-road environments: Build an intelligent obstacle recognition model based on scene scale and DCNN transfer learning. Utilize public datasets and DCNN transfer learning strategies to use scenes as primitives to represent the overall characteristics of obstacles and construct complex semantic patterns. Interpretation of ground mechanics parameters in off-road environments: Construct a rock and soil classification system for ground mechanical properties, survey and identify ground rock and soil characteristic types, and construct ground soil mechanical property parameters for off-road environments based on the soil moisture inversion method based on remote sensing data.

7. The vehicle maneuverability assessment system for off-road maneuvers according to claim 5, characterized in that: In the module M2: Establish a knowledge base of the corresponding mechanical properties of known soil types, obtain a coupled quantitative calculation method that meets the ground mechanical property parameters of different passing vehicles and standard classifications, and realize the inversion of ground mechanical property parameters in unknown areas and the evaluation of vehicle traffic performance.

8. The vehicle maneuverability assessment system for off-road maneuvers according to claim 5, characterized in that: In the module M3: The vehicle-ground interaction model is determined by the type of vehicle and ground. Different types of ground have different characterization parameters. The type of vehicle is determined by the type of its running mechanism, and the type of ground exists in many different ways due to its composition. Based on the interaction mechanism between different walking mechanisms and the ground, a multi-degree-of-freedom mechanical model of the vehicle in a three-dimensional unstructured environment is established, providing a model basis for the evaluation of vehicle passability and maneuverability.

Images