ADAS simulation method for intelligent driving automobileTechnical Field
The invention relates to the technical field of plate processing equipment, in particular to an ADAS simulation method for intelligent driving of an automobile.
Background
With the continuous progress of technology in the automobile industry, people put higher and higher requirements on the safety and the comfort of automobiles, advanced Driving Assistance Systems (ADAS) are widely applied and continuously developed in the automobile industry, and with the rising of intelligent automobiles, the attention to safety problems is increasingly increased, and as automobiles have complex movement characteristics, the automobiles face larger challenges and difficulties in the testing process; with the continuous updating of ADAS and intelligent driving systems, various sensor functions are continuously optimized and perfected, and the interaction between different hardware and the complex association between systems are also increasing. Because the test period is long and ADAS cannot be optimized quickly, the test work becomes time-consuming and labor-consuming in the processes of hardware upgrade and system update.
Currently, in general, an ADAS test for intelligently driving an automobile mainly tests a test vehicle in a test site and records running data of the vehicle, however, the existing whole vehicle test platform can only acquire data in an actual test and perform simple analysis, for example:
Lack of data processing systemization: the test data acquired by the existing platform is simply analyzed, and a deep data processing flow and algorithm are lacked, so that a large amount of complex test data cannot be effectively mined and analyzed. This results in an inability to provide sufficiently detailed and comprehensive data support in discovering problems or optimizing the system.
The function is single: the platform has limited functions, generally can only perform basic data acquisition, storage and simple analysis, and lacks advanced functions such as data model establishment, real-time monitoring, automatic report generation and the like, which are necessary for complex intelligent driving system testing.
The product test requirements cannot be met: due to the lack of systematic processing and diversification functions, existing platforms often fail to meet the comprehensive testing requirements of modern intelligent automobiles. For example, requirements in large-scale data comparison, real-time performance monitoring, multi-scenario simulation, etc. cannot be effectively supported;
therefore, the existing whole vehicle test platform can only acquire data in an actual test and perform simple analysis, lacks systematic processing, has single function and cannot meet the requirement of product test.
Therefore, a person skilled in the art provides an ADAS simulation method for intelligent driving of an automobile, so as to solve the problems set forth in the background art.
Disclosure of Invention
(One) solving the technical problems
Aiming at the defects of the prior art, the invention provides the ADAS simulation method for the intelligent driving automobile, which has the advantages that the actual scene test is carried out once to repeatedly simulate and reproduce the intelligent driving and ADAS simulation system according to the test requirement, the test efficiency of the intelligent driving and ADAS simulation system can be improved, the test cost is reduced and the like, and the problems that a common whole automobile test platform can only acquire data for analysis in the state that an actual measurement automobile participates in the test, the data lacks systematic processing, the function is relatively single, and the test requirement of a product cannot be met are solved.
(II) technical scheme
In order to realize the purpose that the intelligent driving and ADAS simulation system can be repeatedly simulated and reproduced according to the test requirement by using the actual scene acquisition data only by performing one-time actual scene test, the test efficiency of the intelligent driving and ADAS simulation system can be improved, and the test cost is reduced, the invention provides the following technical scheme: an ADAS simulation method for intelligently driving an automobile comprises demand analysis and scene definition, sensor simulation development, environment modeling and generation, vehicle dynamics model development, control algorithm and reality optimization, simulation operation and evaluation and result analysis and optimization;
When the environment modeling and the generation are used for testing the actual scene, the basic related data of intelligent driving and ADAS simulation of the test vehicle needs to be collected, classified and recorded, the original basic data is processed and converted according to specific standards or specifications so as to be used or analyzed later, and finally, the environment data model with the same intelligent driving automobile and ADAS control logic is built.
Preferably, the simulation running and evaluating and the result analyzing and optimizing include: model integration and preparation: integrating a developed sensor model, a vehicle dynamics model, an environment model and a control algorithm into a simulation platform, preparing a scene model, basic data and simulation parameter setting required by simulation, setting and executing the scene, setting a plurality of different driving scenes and test cases, covering various road conditions, traffic conditions and weather conditions, running the scenes in the simulation environment, and observing how an ADAS system responds and executes;
data collection and recording: the data in the simulation process, including sensor data, vehicle state and control input, is recorded in real time, so that the data recording is complete and accurate for subsequent analysis and evaluation, wherein the data optimization module calls the data comparison module to carry out comparison processing, and the data iteration processing can be carried out on the parameters of the hardware by detecting the upgrading condition of the hardware.
Preferably, in the requirement analysis and scene definition, functional requirements of the ADAS system, such as automatic emergency braking and lane keeping assistance, need to be determined, wherein simulation scenes are defined, including road types, traffic flows and weather conditions.
Preferably, the vehicle dynamics model development generally includes basic physical equations describing vehicle motion states and road geometry modeling, specifically, vehicle motion states include the following formulas:
Acceleration (a (t) = \frac { f_ { \text { total } (t) } { m }
Where (F_text { total } } (t)) is the total force acting on the vehicle and (m) is the mass of the vehicle.
Speed (v (t) =v_0+_int_ {0} { t } a (\tau), d\tau)
Where (v_0) is the initial speed.
Position (x (t) =x_0+_int_0 } { t } v (\tau), d\tau)
Wherein (x_0) is the initial position;
Road geometry modeling involves describing the curvature of a road, which can be deduced and established based on the actual shape and design parameters of the road, specifically, the specific formulas are as follows:
The road curvature describes the degree of curvature of the road centerline at any point, typically, the curvature (\kappa (s)) is a function of distance(s), where(s) represents the arc length or straight line distance along the road centerline. Mathematically, the curvature can be deduced from the actual geographical data by parametric equations of the road or by numerical methods. The expression includes:
[\kappa(s)＝\frac{d\theta(s)}{ds}]
where (\theta (s)) is the direction angle of the road centerline at distance(s).
Preferably, in the simulation operation and evaluation and result analysis and optimization, the scene model needs to be recharged to the intelligent driving and ADAS of the test vehicle, the decision making system of the test vehicle makes a decision according to the recharging scene model, and the execution system of the test vehicle executes the decision.
Preferably, in the implementation and optimization of the control algorithm, an ADAS control algorithm, such as brake control, path planning, etc., needs to be developed and implemented, while an optimization algorithm can ensure that a safe and effective response can be provided in various scenarios.
Preferably, the model for developing the vehicle sensing sensor in the development of the sensor model comprises a camera, a laser radar and a millimeter wave radar, so that the model can accurately simulate the working condition of the sensor under different conditions, such as a visual field range, resolution and noise.
Preferably, the environment modeling and generation is primarily for creating a virtual road environment, including road geometry, traffic signs and road markings, and real environmental factors integrating weather, lighting and road conditions.
(III) beneficial effects
Compared with the prior art, the invention provides an ADAS simulation method for intelligent driving of an automobile, which has the following beneficial effects:
1. According to the ADAS simulation method for the intelligent driving automobile, data are collected and recorded in real time, normalized processing and optimization are performed, the integrity and the accuracy of the data are guaranteed, the performance of the intelligent driving system can be effectively analyzed and evaluated, multiple actual scenes can be simulated in a simulation environment through a large number of playback training, the requirements of actual road tests are reduced, the test cost and the time cost are reduced, repeated tests are performed in the simulation environment, and the safety and the reliability of the intelligent driving system are improved.
2. According to the ADAS simulation method for the intelligent driving automobile, the algorithm and the hardware configuration of the intelligent driving system can be optimized through the iterative processing of the data model and the hardware parameters, the system performance and the response capacity are improved, analysis and comparison processing are carried out based on a large amount of simulation data, scientific decision making and system optimization strategies are facilitated, and the development and application level of the intelligent driving technology is further improved.
Drawings
Fig. 1 is a schematic diagram of a simulation system in an ADAS simulation method for driving an automobile intelligently.
Detailed Description
The following will clearly and completely describe the technical solutions in the embodiments of the present invention, and it is obvious that the described embodiments are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, an ADAS simulation method for intelligent driving of an automobile includes demand analysis and scene definition, sensor simulation development, environment modeling and generation, vehicle dynamics model development, control algorithm and reality optimization, simulation running and evaluation, and result analysis and optimization, where functional demands of an ADAS system need to be determined, such as automatic emergency braking and lane keeping assistance, where simulation scenes including road types, traffic flows, and weather conditions are defined, and environment modeling and generation are mainly used to create virtual road environments including road geometry, traffic signs, and road markings, and real environmental factors integrating weather, illumination, and road conditions;
When the environment modeling and the generation are used for testing an actual scene, basic related data of intelligent driving and ADAS simulation of a test vehicle are required to be acquired, classified and recorded, the original basic data are processed and converted according to specific standards or specifications so as to be used or analyzed later, finally, an environment data model with the same intelligent driving automobile and ADAS control logic is established, and models of vehicle sensing sensors, including cameras, laser radars and millimeter wave radars, are developed in sensor model development, so that the models can accurately simulate the working conditions of the sensors under different conditions, such as visual field range, resolution, noise and the like;
The simulation running and evaluation in the simulation running and evaluation and result analysis and optimization comprises the following steps: model integration and preparation: integrating a developed sensor model, a vehicle dynamics model, an environment model and a control algorithm into a simulation platform, preparing a scene model, basic data and simulation parameter setting required by simulation, setting and executing the scene, setting a plurality of different driving scenes and test cases, covering various road conditions, traffic conditions and weather conditions, operating the scenes in the simulation environment, observing how an ADAS system responds and executes, and in simulation operation and evaluation and result analysis and optimization, recharging the scene model to intelligent driving and ADAS of a test vehicle, making a decision by a test vehicle decision system according to the recharging scene model, and executing the decision by the test vehicle execution system;
Data collection and recording: the data in the simulation process, including sensor data, vehicle state and control input, is recorded in real time, so that the data recording is complete and accurate for subsequent analysis and evaluation, wherein the data optimization module calls the data comparison module to carry out comparison processing, and the data iteration processing can be carried out on the parameters of the hardware by detecting the upgrading condition of the hardware;
vehicle dynamics model development typically includes basic physical equations describing vehicle motion states and road geometry modeling, specifically, vehicle motion states include the following formulas:
Acceleration (a (t) = \frac { f_ { \text { total } (t) } { m }
Where (F_text { total } } (t)) is the total force acting on the vehicle and (m) is the mass of the vehicle.
Speed (v (t) =v_0+_int_ {0} { t } a (\tau), d\tau)
Where (v_0) is the initial speed.
Position (x (t) =x_0+_int_0 } { t } v (\tau), d\tau)
Wherein (x_0) is the initial position;
Road geometry modeling involves describing the curvature of a road, which can be deduced and established based on the actual shape and design parameters of the road, specifically, the specific formulas are as follows:
The road curvature describes the degree of curvature of the road centerline at any point, typically, the curvature (\kappa (s)) is a function of distance(s), where(s) represents the arc length or straight line distance along the road centerline. Mathematically, the curvature can be deduced from the actual geographical data by parametric equations of the road or by numerical methods. The expression includes:
[\kappa(s)＝\frac{d\theta(s)}{ds}]
Wherein (\theta (s)) is the direction angle of the road centerline at distance(s);
In the implementation and optimization of control algorithms, it is necessary to develop and implement ADAS control algorithms, such as brake control, path planning, etc., while optimization algorithms can ensure that safe and effective response is provided in various scenarios.
In summary, when the test vehicle tests an actual scene, various sensors record data of the vehicle in an actual driving process, including scene information, environment state, sensor data, lane information, vehicle model, real-time state and the like, and the data is classified, recorded and processed, and then a control system sends out an instruction to optimize the data, establish a data model, encrypt and store the data. Meanwhile, a software GUI graphical user interface is established, and the acquired data is transmitted to a control system after coupling and filtering processing. The control system and the data processing module establish data communication through a network protocol, data are collected in real time by the data collecting and storing module, a data model is established by the data processing module and recorded to a database, then the collected data are processed regularly and sent to the data optimizing module, then the data are compared and processed through the data comparing module, iteration processing is carried out on hardware parameters, finally, a large number of playback training is carried out on the test vehicle by utilizing the data model, the test efficiency of the intelligent driving and ADAS simulation system is improved, the test cost is reduced, the data are collected and recorded in real time, the normalization processing and optimization are carried out, the integrity and the accuracy of the data are ensured, the performance of the intelligent driving system is effectively analyzed and evaluated, a plurality of actual scenes can be simulated in the simulation environment by a large number of playback training, the requirement of actual road tests is reduced, the test cost and the time cost are reduced, repeated tests are carried out in the simulation environment for a plurality of times, the potential safety problem and system defects are found, the safety and the reliability of the intelligent driving system are improved, the iteration processing of the data model and the hardware parameters is improved, the intelligent driving system can be further improved, the intelligent driving system is improved, the performance and the intelligent driving system is further improved, the intelligent driving system is applied to the decision-making and the intelligent driving system is more based on the intelligent driving system, and the intelligent driving system is optimized, and the application strategy and the intelligent driving system is improved, and the performance and the intelligent driving system is more scientific and improved.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.