CN113962019A - Intelligent driving automobile safety protection system based on virtual reality technology - Google Patents

Abstract

The invention relates to an intelligent driving safety protection system based on virtual reality, which comprises a software simulation module, a hardware design and platform building module, a computer control module and a safety protection control strategy module, wherein the software simulation module is connected with the hardware design and platform building module, the software simulation module and the hardware design and platform building module are connected with the computer control module, and the computer control module is connected with the safety protection control strategy module. From the perspective of safety protection, a hardware-in-loop test is carried out in a mode of combining software and hardware, the state information of the vehicle is monitored truly, the safety protection strategy is applied to the real vehicle, dangerous scenes can be reproduced by depending on a test platform, the efficiency of testing and verifying the safety protection strategy is improved, and the safety of intelligent driving is improved greatly.

Description

Intelligent driving automobile safety protection system based on virtual reality technology

Technical Field

The invention relates to the technical field of intelligent automobiles and virtual reality, in particular to an intelligent driving automobile safety protection system based on a virtual reality technology.

Background

With the continuous development of the intelligent driving automobile technology, the complexity of the corresponding verification test is increased. At present, the intelligent driving safety problem caused by equipment failure, environmental interference, misoperation of a driver, driving fatigue, dangerous collision and the like is particularly prominent, and at the moment, an automobile driving system needs to respond in time and execute active safety operation when the problems occur, so that the driving safety of an automobile is ensured. In the related technology of intelligent driving of automobiles, the research and development of the safety protection function of an intelligent automobile are particularly important, but the problems of safety, cost and the like exist in the actual road environment and the actual automobile test.

Virtual Reality (VR) is also called Virtual environment, and is a technology for building a Virtual space by applying computer graphics and simulation technology. The technology integrates the latest development of technologies such as computer graphics, computer simulation, artificial intelligence, induction, display, network parallel processing and the like, is a simulation system generated by assistance of computer technology, and has the characteristics of immersion, interactivity, imagination and the like.

With the vigorous development of global virtual reality technology, the VR technology in the automobile industry is increasingly paid attention to in research and development, and the automobile driving simulator based on the virtual reality technology is an important tool in the research and development of intelligent driving technology and plays an important role in the research and development of the intelligent driving technology. The related technology of intelligent driving needs to go through a large amount of road tests before being applied to the actual commercialization to meet the commercial requirements. As a new thing, the development of intelligent driving automobiles still faces a lot of problems, such as time cost of road test, requirements of automatic driving laws and regulations of each country, safety of dangerous working conditions and extreme working condition scene tests, difference between road traffic environment and design requirements of each country, and the like, and all bring many difficulties to research and development tests of intelligent driving related technologies.

According to related research, the intelligent driving automobile algorithm at least needs to accumulate 177 hundred million kilometers of driving data to reach the level of human drivers. If 100 test vehicles are used and the road test is continuously carried out 24 hours a day, and the average speed per hour is 40 kilometers, more than 500 years are needed to complete the target mileage, and the consumed time and manpower and material resources are difficult to bear. Dangerous working condition and extreme working condition scenes are key parts of verification of related technologies of intelligent driving, and are necessary test scenes for testing and verifying effectiveness, reliability and safety of the related technologies of the intelligent driving, but the extreme scenes can be unavailable in reality, and the real vehicle test risk is also large. At present, absolute safety cannot be guaranteed for intelligent driving automobiles, and China governments keep a careful attitude for testing open intelligent driving automobile roads. In addition, the lack of corresponding traffic laws and regulations and insurance claim settlement mechanisms in the field of China also restricts the development of large-scale open road tests of intelligent driving automobiles. Only by means of partial open roads and intelligent networking test areas for road test, the ever-increasing test requirements of the intelligent driving automobile are difficult to meet, and the global development and the technical communication of the industrial chain of the intelligent driving automobile face numerous practical problems.

At present, a virtual reality technology is applied to intelligent driving, and the virtual vehicle is controlled to accelerate, decelerate, shift gears, brake, overtake or switch lanes and the like in a virtual environment based on dynamic attributes of a real environment vehicle, namely, simulation of a traffic scene is usually emphasized, some technologies are designed to perform simulation analysis aiming at the operation habit of a driver or build a vision system, but a control strategy and a solution method when safety threats such as severe environment, fatigue of the driver, equipment failure and the like occur are not considered, and the problems faced by intelligent driving are not solved from the perspective of safety protection.

The simulation test of the intelligent driving related technology by using the automobile driving simulator based on the virtual reality technology is an efficient route for solving the challenge of intelligent driving research and development test. In addition, china is the largest automobile producing and consuming country in the world, and a simulation test based on virtual reality is one of key core technologies in the research and development process of intelligent driving automobiles, and independent research and development must be realized to take a leading position in international competition.

The existing application of the virtual reality technology to intelligent driving has limitations, and the role of the virtual reality technology in intelligent driving is not systematically researched from the perspective of safety protection. The development of the intelligent driving technology brings many difficulties to the research and development of the intelligent driving technology in the aspects of time cost of road testing, safety of dangerous working conditions and extreme working condition scene testing, difference of road traffic environment and design requirements and the like.

Disclosure of Invention

The invention provides an intelligent driving automobile safety protection system based on a virtual reality technology, aiming at solving the safety problems caused by equipment faults, environmental interference, misoperation of a driver, driving fatigue, dangerous collision and the like in the existing intelligent driving.

The technical scheme of the invention is as follows:

an intelligent driving safety protection system based on virtual reality is characterized by comprising a software simulation module, a hardware design and platform building module, a computer control module and a safety protection control strategy module, wherein the software simulation module is connected with the hardware design and platform building module;

the software simulation module comprises a vehicle dynamics modeling unit, a terrain modeling and visual simulation rendering unit and a driving simulation interaction unit, wherein the vehicle dynamics modeling unit analyzes a vehicle dynamics system according to an object-oriented theory, sets the matching data of the performance parameters of the whole vehicle, establishes a vehicle dynamics model based on virtual reality and realizes the motion simulation of the vehicle in a virtual prototype; the terrain modeling and visual simulation rendering unit establishes a three-dimensional high-definition terrain scene model based on a three-dimensional modeling technology aiming at safety threats in intelligent driving, and processes the three-dimensional high-definition terrain scene model through a real-time simulation program, an image generator and a visual effect database so as to realize virtual driving visual simulation rendering of the automobile; the driving simulation interaction unit utilizes a virtual reality technology to design an automobile driving virtual simulation experiment, and carries out automobile driving dynamic simulation, collision detection and multi-view switching;

the hardware design and platform building module comprises a hardware test platform design unit and a real-time data acquisition unit, wherein the hardware test platform design unit is used for designing a multifunctional multi-degree-of-freedom driving simulation test platform with human-vehicle-environment interaction characteristics, simulating a moving part and an operating mechanism of a cockpit, and realizing simulation of realistic tactile feeling in the driving process of a real vehicle in the cockpit; the real-time data acquisition unit is used for acquiring the operation data of a driver in real time so as to judge the state of the driver and the current state of the vehicle and realize effective real-time communication between the output equipment and the simulation;

the computer control module receives vehicle attitude signals transmitted from the terrain modeling and visual simulation rendering unit, and the vehicle attitude signals are converted into corresponding analog quantity control driving simulation test platforms through calculation, so that the man-machine interaction simulation attitude in the cockpit is realized, safety protection control strategies and algorithms are verified and optimized, and the danger early warning threshold value and the driving strategy set by the vehicle are adaptively adjusted;

the safety protection control strategy module makes a safety protection control strategy and an optimization algorithm aiming at safety threats existing in intelligent driving, and a vehicle is designed to adopt corresponding safety operation in a corresponding scene, so that active safety and auxiliary driving of intelligent driving are realized.

Preferably, the vehicle dynamics modeling unit adopts a simulation platform and virtual simulation software to establish a vehicle dynamics model based on virtual reality, the simulation platform adopts a Simulink simulation platform, and the virtual simulation software adopts Carsim multi-body dynamics simulation software.

Preferably, the safety threats in the terrain modeling and vision simulation rendering unit include at least one of equipment failure, system function limitations, personnel misoperation, dangerous collision and special environmental interference.

Preferably, the driving simulation test platform in the hardware test platform design unit comprises a steering wheel, an accelerator, a brake, a dynamic seat, an induction helmet and a simulation display, and has three-dimensional display, position capture and head tracking functions.

Preferably, the manipulation data in the real-time data collection unit includes at least one of an opening degree of an accelerator, a steering wheel angle, a brake pedal angle, and a gear position.

Preferably, the hardware design and platform building module further comprises a hardware platform function verification unit, wherein the hardware platform function verification unit is combined with the real-time data acquisition unit and used for verifying the consistency, consistency and accuracy of the simulation rendering of the driving simulation test platform and the virtual driving visual scene of the automobile.

Preferably, the computer control module further performs I/O channel design, interface circuit design, anti-interference capability design and human-computer interaction interface design.

Preferably, the safety protection control strategy module comprises at least one of a designated driver state monitoring strategy, a driver misoperation protection strategy, a severe environment protection strategy and an obstacle collision avoidance protection strategy; when the driver misoperation or the operation exceeding the danger threshold value is detected, the vehicle is informed to give a prompt warning, and the vehicle is controlled to adopt active braking and active steering operation according to the real-time state; when the severe environment interferes with normal driving, the vehicle is informed to respond and prompt warning in time, and the vehicle is controlled to adopt active braking and active steering operation according to the real-time state; and designing a control algorithm when vehicle collision is simulated, and informing the vehicle to safely avoid collision according to dangerous conditions.

Preferably, the terrain modeling and view simulation rendering unit adopts a Multigen Creator to build a three-dimensional high-definition terrain scene model.

Preferably, the computer control module adopts a PC bus structure.

The invention has the beneficial effects that:

the invention relates to an intelligent driving safety protection system based on virtual reality, which is provided with a software simulation module, a hardware design and platform building module, a computer control module and a safety protection control strategy module, wherein a safety dangerous scene in intelligent driving is reproduced through the software simulation module, the hardware design and platform building module, after a dangerous condition occurs, the safety protection control strategy module sends an active safety signal to the hardware design and platform building module through the computer control module according to a control strategy, and sends a control instruction to equipment such as a steering wheel, a seat and the like to ensure the safety of a driver, and under the condition, a real vehicle state change is fed back to the computer control module to form closed-loop feedback control, so that the vehicle reaches a safe state, and the provided safety protection strategy is tested and verified. The system aims at the safety threat in intelligent driving, the virtual reality technology is applied to intelligent driving safety protection, a safety protection system with the interactive characteristic of human-vehicle-environment and high operation precision and good stability and real-time performance is researched and developed, hardware in-loop test is carried out in a software and hardware combination mode from the safety protection angle, the state information of a vehicle is monitored truly, a safety protection strategy is applied to a real vehicle, multiple safety threat test environments are reproduced, active safety measures are taken, dangerous scenes can be reproduced by depending on a test platform according to formulated control strategies and optimization algorithms, the efficiency of testing and verifying the safety protection strategy is improved, and the safety of intelligent driving is greatly improved.

Drawings

Fig. 1 is a schematic structural diagram of an intelligent driving safety protection system based on virtual reality according to the present invention.

Fig. 2 is a schematic flow diagram of the intelligent driving safety protection system based on virtual reality according to the present invention.

Detailed Description

For a clearer understanding of the contents of the present invention, reference will be made to the accompanying drawings and examples.

Aiming at the safety problems caused by equipment faults, environmental interference, misoperation of a driver, driving fatigue, dangerous collision and the like in intelligent driving, the invention develops an intelligent driving safety protection system based on a virtual reality technology, develops a safety protection system with high operation precision, good stability and good real-time performance and the interactive characteristic of human-vehicle-environment according to a formulated control strategy and an optimization algorithm, reproduces various safety threat test environments and takes active safety measures, improves the active safety of intelligent driving according to the formulated control strategy and the optimization algorithm, and can reduce the test cost and improve the test efficiency. The system has important significance in applying the virtual reality technology to the research and development of the intelligent driving technology.

The intelligent driving safety protection system based on virtual reality embodies important advantages gradually in the aspect of testing of automobile intelligent driving related technologies, and not only the test efficiency is higher, but also the intelligent driving safety protection system is easy to realize, and specifically as follows: testing dangerous working conditions and extreme working conditions. Some technologies of intelligent driving need to be tested under extreme working conditions, and the safety of real vehicle tests cannot be guaranteed. The virtual reality technology can simulate the scenes of dangerous working conditions and extreme working conditions without any risk, and remind a driver to carry out active safety operation, so that the intelligent driving function can be conveniently and reliably researched, and the safety and convenience of intelligent driving automobile testing can be greatly improved. Secondly, various test simulation working conditions and simulation scenes can be repeated, and if the test simulation working conditions and the simulation scenes which can be repeatedly tested are difficult to obtain continuously in the real vehicle environment, the test simulation working conditions comprise extreme working conditions and dangerous working conditions. Time cost and real vehicle test cost, a large number of experimental conditions are needed to support in the development process of the intelligent driving vehicle, and all test work is concentrated on a real vehicle, so that the test scheme is long in test period and extremely high in cost. Different from the common simulation test, the automobile driving simulation system with the motion platform can test the subjective feeling of people who intelligently drive various algorithms under the safe and repeatable conditions, so that the method has guiding significance on the improvement of the various algorithms and is more targeted.

The invention relates to an intelligent driving safety protection system based on virtual reality, which is structurally shown in figure 1 and comprises a software simulation module, a hardware design and platform building module, a computer control module and a safety protection control strategy module, wherein the software simulation module is in soft connection with the hardware design and platform building module through communication data, the software simulation module and the hardware design and platform building module are connected with the computer control module, and the computer control module is connected with the safety protection control strategy module. The system comprises a software simulation module, a hardware design and platform building module and a hardware platform function verification module, wherein the software simulation module comprises a vehicle dynamics modeling unit, a terrain modeling and visual simulation rendering unit and a driving simulation interaction unit; the computer control module is used for processing the behavior data of the driver, transmitting the sensing equipment data and the road environment data input by the software simulation module; and the safety protection control strategy module is used for solving the dangerous situation of safety threat in intelligent driving, sending a signal to the hardware design and platform building module according to the control strategy, and controlling the hardware equipment of the test platform to carry out active safety operation so as to ensure the safety of a driver.

The specific structure in each module will be described in further detail below.

The software simulation module comprises a vehicle dynamics modeling unit, a terrain modeling and visual simulation rendering unit and a driving simulation interaction unit, and virtual reality-based vehicle dynamics models, three-dimensional high-definition terrain modeling and visual simulation rendering and driving simulation interaction are respectively established.

(1) A vehicle dynamics modeling unit: the vehicle dynamics system is analyzed according to an object-oriented theory, the vehicle system is decomposed into a plurality of subsystems, and a mathematical model which accords with the reality and is suitable for real-time simulation calculation is established on the basis of abstraction and simplification according to the characteristics of the subsystems. And the matching data of the performance parameters of the whole vehicle are set according to different requirements, and a geometric model and a mathematical model of the vehicle are combined, so that the characteristics and the movement characteristics of the real vehicle are met, and a good movement simulation effect of the vehicle in a virtual prototype is realized. A vehicle dynamics model based on virtual reality can be established by respectively adopting an MATLAB/Simulink simulation platform and Carsim multi-body dynamics simulation software. The accuracy of the vehicle dynamics model is an important factor related to the fidelity of the simulation of the simulated driving, and establishing a suitable and effective vehicle dynamics model is one of the key technologies for realizing the whole simulated driving system.

It should be noted that, the use of MATLAB/Simulink simulation platform and multi-body dynamics simulation software such as Carsim to build a vehicle dynamics model based on virtual reality is a preferred embodiment, but not an exclusive implementation, and other simulation platforms and simulation software capable of implementing the building of a vehicle dynamics model based on virtual reality may be used, for example, simulation platforms such as SCILAB, Octave, and AMEsim may also be used, and simulation software such as Prescan, Carmaker, ADAMS, Cruise, and ASM may also be used to build a vehicle dynamics model based on virtual reality.

Specifically, a vehicle dynamics system of which a research object is an automobile simulated driving system is analyzed according to an object-oriented theory, the vehicle dynamics system is decomposed into a plurality of subsystems, the scale of a model and boundary conditions related to the scale are determined according to a specific object and a research target, objects such as a vehicle body, a suspension, wheels, a power train, a transmission system, a steering system and a braking system of the vehicle dynamics system are finally established, an object model, a functional model and a dynamic model of each subsystem are analyzed, coupling and correlation are carried out among all the parts, and a mathematical model which is suitable for real-time simulation calculation is established on the basis of abstracting and simplifying each subsystem. And integrating the motion equation, calculating a new state of the object according to the time variation, researching a function calling mechanism used under MATLAB/Simulink, packaging a solver by using an S-function, and building the solver and a calculation model in the Simulink to realize simulation of a vehicle dynamics model in the Simulink.

It can be understood that, because the vehicle dynamic model includes a plurality of components and some parameters cannot be measured, the established dynamic model cannot ensure the accuracy of the system as a whole, and meanwhile, the complex model has a slow operation speed in the simulation calculation and cannot meet the requirement of real-time simulation. Therefore, a multi-body dynamics simulation software Carsim capable of realizing real-time interaction is selected for vehicle dynamics modeling, a complex multi-body system dynamics equation is deduced and solved, each component of a vehicle system is regarded as a rigid body or an elastic body, motion constraint is adopted to connect each component, the real-time performance is guaranteed while the calculation accuracy is considered, motion parameters including position, speed, acceleration, various angles, angular velocity, angular acceleration and the like are output externally, a developed standard API (application programming interface) interface function programming mode is adopted to perform comparative analysis on simulation calculation of typical operation working conditions, road test results and software simulation results, a vehicle dynamics model with higher accuracy is obtained, and the vehicle dynamics model can be applied to a simulated driving system to perform safety simulation research.

(2) A terrain modeling and visual simulation rendering unit: the method is used for carrying out visual modeling aiming at the safety threat problem existing in intelligent driving, and mainly comprises safety threats caused by electronic equipment faults, system function limitations, personnel misoperation, pedestrian collision avoidance, special environment interference and the like during vehicle driving. A three-dimensional high-definition terrain scene model is established by adopting three-dimensional modeling software, and a basic class library such as a hierarchical structure and model analysis of a virtual driving scene is designed based on a graphic development kit and a programming language for calling the scene. The scene model is optimized through modules such as a real-time simulation program, an image generator, a visual effect database and the like, and an automobile virtual driving scene simulation system is constructed for simulation rendering so as to build a vivid virtual driving environment.

The terrain modeling and view simulation rendering unit is an important component of an intelligent driving safety protection system, a computer real-time image generation system is utilized to generate virtual environments seen by a driver in the driving process of an intelligent automobile, such as traffic signs, trees, buildings, sidewalks, vehicles, pedestrians and the like, and images with certain fidelity are generated in a view simulation submodule in real time along with operation signals input by the driver. The Multigen Creator/Vega product is used as a development tool of the virtual view simulation system. Its development is divided into three main steps: establishing a three-dimensional model, configuring a virtual scene and creating Vega real-time application. And (3) creating an automobile three-dimensional model and a virtual scene in the Creator, realizing real-time driving of the virtual scene in the Vega, and finally finishing development of a virtual scene simulation system.

Specifically, as shown in the flow diagram of fig. 2, aiming at the problem of safety threats (mainly including safety threats caused by electronic equipment faults, system function limitations, personnel misoperation, pedestrian collision avoidance, special environment interference and the like during vehicle driving), a Multigen Creator is adopted to establish a three-dimensional high-definition terrain scene model, basic class libraries such as a hierarchical structure of a virtual driving scene, model analysis, a collision detection technology, control of a motion camera and the like are designed based on a graphic development kit and a Visual C + + programming language, a virtual scene is configured for scene calling, an Open GVS 3D Visual management software is adopted through an object-oriented graphic rendering technology, a three-dimensional graphic rendering library and a three-dimensional model library including mechanisms such as scene management, image rendering, message management and the like are established, real-time driving of the virtual scene is realized in Vega, a high-performance computer image generation system and a multi-type graphic acceleration card are used, the real-time and vivid image processing is realized, the frame rate required by real-time performance is achieved by adopting technologies such as developing parallelism, researching management of virtual scenes, rapidly accessing a scene database, utilizing model segmentation, detail level division, memory management and the like in a three-dimensional modeling technology, the three-dimensional high-definition terrain scene model is optimized by generating a large-range visual effect of immersion through optical display equipment, and finally, an automobile virtual driving scene simulation submodule is constructed to create a real-time and vivid virtual driving environment.

It should be noted that, the establishment of the three-dimensional high-definition terrain scene model by using the Multigen Creator is a preferred embodiment, but not an exclusive embodiment, and other simulation tools capable of realizing establishment of the three-dimensional high-definition terrain scene model may be used, for example, simulation tools such as 3DMax, MAYA, cara, Autodesk, Rhino, Udacity, Sketch up, Deep expansion, and the like may also be used to establish the three-dimensional high-definition terrain scene model.

(3) Driving simulation interaction unit: the virtual simulation experiment of automobile driving is designed by utilizing the virtual reality technology, and dynamic simulation, collision detection and multi-view switching of automobile driving are carried out. That is to say, the virtual driving environment of the automobile created by using the virtual reality technology can simulate the driving training under various weather conditions (such as weather of rainy days, foggy days, snowy days and the like), various road conditions (such as roads of city streets, expressways, mountain roads, concave-convex road surfaces and the like) and various time periods (such as daytime and night), develop the virtual simulation experiment project of the automobile driving based on the virtual reality technology, research the influence of various time periods, various road conditions, various weather and the like on the driving safety under intelligent driving, and realize the all-weather virtual simulation experiment of the automobile driving.

And after the software simulation module is finished, the hardware design and platform building module is used for carrying out hardware design and platform building.

And the hardware design and platform building module comprises a hardware test platform design unit, a real-time data acquisition unit and a hardware platform function verification unit, and is used for respectively carrying out hardware test platform design, real-time data acquisition and hardware platform function verification.

(1) Hardware test platform design unit: the method is used for designing a set of multifunctional multi-degree-of-freedom driving simulation test platform with human-vehicle-environment interaction characteristics, high operation precision, high stability and high real-time performance, can simulate moving parts and an operating mechanism of a driving cabin, and realizes the realistic simulation of the touch feeling in the driving process of a real vehicle in the limited simulated driving cabin by the aid of a virtual scene which dynamically responds according to the actions of a driver. Preferably, the driving simulation test platform comprises equipment such as a steering wheel, an accelerator, a brake, a dynamic seat, an induction helmet and a simulation display, and has the functions of three-dimensional display, position capture, head tracking and the like. The main control computer with good performance and strong graphic processing capability is adopted to generate a real-time simulation visual scene, and the real-time simulation visual scene is communicated with an operating system and a motion simulation control system in real time. The motion simulation control system adopts a multifunctional multi-degree-of-freedom shaking device, the shaking device adopts a servo motor which has the characteristics of high sensitivity, small null shift, wide frequency band, strong anti-pollution capability and the like, and also adopts a QDY electro-hydraulic servo valve which has the advantages of high reliability and high efficiency and can efficiently output stable rated pressure, flow and current, and adopts a displacement sensor, an oscillator, a phase-sensitive demodulator and a differential transformer, wherein the displacement sensor adopts a direct current differential transformer type displacement sensor, the oscillator, the phase-sensitive demodulator and the differential transformer are packaged together by the direct current differential transformer type displacement sensor, and only a stable direct current power supply is required to be provided, so that direct current output voltage which is in linear relation with displacement can be obtained, and the voltage output signal is larger.

(2) A real-time data acquisition unit: the real-time monitoring system is used for collecting the operation data of the driver in real time so as to judge the state of the driver and the current state of the vehicle and realize effective real-time communication between the output equipment and the simulation. The signal acquisition can be carried out by adopting a timer with higher transmission precision, a digital quantity data acquisition card and an analog quantity data acquisition card. The driver carries out subjective judgment and makes operation decision according to the provided traffic visual scene, sends out operation instructions through a steering wheel and various operation mechanisms (mechanisms such as a gear shifting knob, a brake pedal, an accelerator pedal, a speed changing handle, a starting switch and other circuit switches), acquires operation data such as the opening degree of an accelerator, the turning angle of the steering wheel, the angle of the brake pedal, the position of a gear lever and the like through a sensor with an operation touch simulation function, transmits the operation data to a central control computer through a transmission device, and is used as vehicle dynamics system input together with vehicle parameters to calculate the current working condition and posture data of the vehicle, so that high-speed effective real-time communication is realized between output equipment and simulation, more real visual experience is obtained, and the system has stronger initiative, immersion and interactivity.

Preferably, the central control computer of the automobile driving simulation system adopts a Windows operating system, and the driving simulation system is used for receiving the vision simulation submodule data, calculating the vision simulation submodule data and controlling the electro-hydraulic servo system. Therefore, PC hardware resources which can meet the installation requirements are selected, a larger digital screen with higher resolution and a picture ratio of 16:9 is used as visual display equipment, and the visual display equipment is installed on a simulated driving test platform to obtain a better visual effect.

(3) The hardware platform function verification unit is combined with the real-time data acquisition unit, and in a virtual environment, according to a dynamic driving vehicle dynamic model, the dynamic animation system comprises a multi-angle switching function, a sound effect simulation function and a data management function, and is designed to generate dynamic animation in real time, so that a driver is immersed in the virtual driving environment. The vehicle dynamic model is placed in a terrain modeling and visual simulation rendering unit scene, the scene refers to an intelligent driving safety threat scene caused by equipment faults, environmental interference, misoperation of a driver, driving fatigue and dangerous collision, and based on a virtual reality technology, vehicle visual angles can be switched at multiple angles, including front view, back view, side view and overlook, and the change of the vehicle state is observed at multiple angles. The driver can transmit real-time action signals to the terrain modeling and visual simulation module by operating the square disc, the accelerator and the brake pedal, and feed back the change of the vehicle dynamic model to the actual driver, the vehicle body state change of the vehicle is fed back by the multi-degree-of-freedom shaking device, and the environment and the special sound of the vehicle can be simulated by the sound effect simulation function so as to achieve a vivid simulation environment, so that the driver is immersed in the virtual driving environment, the feedback process of-driver operation-vehicle model change-environment feedback-vehicle state under the security threat scene is realized, and the consistency, consistency and accuracy of the simulation rendering of the driving simulation test platform and the virtual driving visual simulation of the vehicle are verified.

And the computer control module is used as a neural center for connecting the terrain modeling and visual simulation rendering unit and the hardware test platform design unit, firstly receives a vehicle attitude signal transmitted from the terrain modeling and visual simulation rendering unit, converts the vehicle attitude signal into a corresponding analog quantity through certain calculation to control the driving simulation test platform, realizes the man-machine interaction analog attitude in the cockpit, verifies and optimizes a safety protection control strategy and algorithm, and adaptively adjusts a danger early warning threshold value and a driving strategy set by the vehicle.

The virtual reality technology is utilized to create an automobile driving environment and simulate various weather conditions such as rainy days, foggy days, snowy days and the like; under various road conditions, such as urban streets, expressways, mountain roads, concave-convex road surfaces and the like; driving training for multiple time periods such as daytime and nighttime, etc.; the method develops an automobile driving simulation experiment scene based on a virtual reality technology, researches the influence of factors such as various time periods, various road conditions and various weather on driving safety under intelligent driving, and realizes all-weather automobile driving virtual simulation experiments. Real-time data in human-computer interaction is monitored and visualized, a series of acquired data including sensing equipment data, road condition data, driving behavior data, traffic and map information data and the like are processed, analyzed and mined by a computer control module, real-time observation can be achieved, algorithm verification, optimization and the like are conducted on the basis of the big data, and safety problems of ubiquitous vehicle sudden conditions such as electronic appliance faults, personnel misoperation, pedestrian collision avoidance, special environment interference and the like are researched and result analysis is conducted. The control strategy and algorithm of the safety protection system are verified and optimized, the danger early warning threshold value can be set for the vehicle, the driving strategy can be adaptively adjusted, and a larger data base and source are provided for vehicle driving and road traffic. The computer control module connects all the scattered subsystems together, tests are carried out on the running state of the vehicle, the security threat scene, the man-machine interaction operation and the like, and the consistency, the consistency and the accuracy of the test bed and the design system are verified.

Because the control precision required by the mechanism is higher, and the control unit is more complex, the control is carried out by adopting a computer. The driving simulation test platform comprises a plurality of hydraulic devices, each hydraulic device is controlled by a servo control system, the servo control systems adopt distributed control and comprise servo cylinders and valve control hydraulic cylinders, the displacement instruction calculation of each servo cylinder is completed by a microcomputer, the position of a mechanism is monitored, one single chip microcomputer controls one valve control hydraulic cylinder, the controller is flexibly changed by adopting a soft servo mode, and only reprogramming is needed, so that one algorithm can be replaced, and complex compensation and control algorithms can be realized on the computer according to requirements.

(1) Bus selection: the computer control module adopts a supervisory control system, the upper computer adopts a high-performance industrial personal computer and adopts a PC bus structure, so that the hardware design is simplified, a user can directly select a functional template meeting the bus standard according to the requirement without considering the matching problem of template plug-ins, and the hardware design of the system is greatly simplified.

(2) I/O channel design: the method is used for designing various input and output templates, process parameters are generally non-electric, and are required to be converted into equivalent electric signals through a sensor, so that input and output channels are required to be adopted for conversion, and the conversion generally comprises an analog input channel, an analog output channel, a digital input channel and a digital output channel.

(3) Designing an interface circuit: the design of universal interface circuit for microcomputer is that the external equipment and process channel can not be directly controlled by host computer, and the microcomputer interface circuit can be used for transferring correspondent information and command, and the actually-used interface is formed from serial interface, parallel interface and management interface (including interrupt management, direct access DMA management and counting/timing).

(4) Designing the anti-interference capability of a computer: in the operation of the computer control module, high and low frequency electromagnetic interference and large instantaneous voltage fluctuation caused by high voltage, strong current, contact sparks when large-scale electric equipment is frequently started and stopped and the like directly influence the normal operation of the computer, so that the high anti-interference capability of the computer control module is effectively designed.

(5) Designing a human-computer interaction interface: real-time data in human-computer interaction are monitored and visualized, the intelligent automobile processes, analyzes and excavates a series of acquired data, real-time observation can be conducted, and research and result analysis can be conducted on ubiquitous safety problems such as electronic appliance faults, personnel misoperation, pedestrian collision avoidance and special environment interference.

And fourthly, a safety protection control strategy module, which is used for formulating a safety protection control strategy and selecting a proper optimization algorithm aiming at safety problems caused by electronic equipment faults, environmental interference, misoperation of a driver, driving fatigue, dangerous collision and the like in intelligent driving, designing a vehicle to adopt safety operation in a corresponding scene, and realizing active safety and driving assistance of intelligent driving. The vehicle gives a prompt warning by sensing the fatigue and distraction states of the driver, and controls the vehicle to adopt active braking and active steering operation according to the running state of the vehicle. Judging the driving operation of a driver, setting a danger threshold value, when detecting that the misoperation and operation of the driver exceed the danger threshold value, the vehicle gives out a prompt warning, and controls the vehicle to adopt active braking and active steering operation according to the real-time state. The simulation natural environment is rendered through visual simulation, when the environment is interfered to normal driving, the vehicle timely makes a response to prompt and warn a driver, and the vehicle is controlled to adopt active braking and active steering operation according to real-time vehicle conditions. When a pedestrian collides with a vehicle in a simulation mode, a control algorithm is designed, and the vehicle can realize safe collision avoidance according to dangerous conditions. According to the environment perception, when the influence of the blind area of the visual field on normal driving of the vehicle is judged, a danger prompt warning is timely made, and the braking, accelerating and steering operations of a driver are properly interfered by using an automobile stabilizing system, so that the driving safety of the vehicle is ensured.

The specific protection process is as follows:

driver state monitoring function: the device comprises a camera, a driver action sensing device, an active early warning device and the like, which are arranged on a simulated driving test platform, wherein the distance between a front vehicle and a self vehicle and the distance between a side lane vehicle and the self vehicle are calculated through a virtual driving vision simulation submodule, so that whether the driver has states of fatigue, attention transfer and the like can be detected, once the vehicle is detected to be in an unsafe state or the driver is detected to be in a state of poor consciousness, sound and visual alarm is provided, and the vehicle is controlled to adopt active braking and active steering operation according to the current driving state of the vehicle.

Driver misoperation safeguard function: the camera and each actuator are used for operating the sensor to monitor and judge the driving operation of a driver, a corresponding danger threshold value is set according to the driving state of the vehicle, when the fact that the misoperation and operation of the driver exceed the danger threshold value are detected, the vehicle can give a corresponding prompt warning, and then the vehicle is controlled to adopt active braking and active steering operation through algorithm optimization according to the real-time state.

The active protection function under the complex severe environment is as follows: the simulation natural environment is rendered by utilizing visual simulation, the simulation natural environment comprises severe weather environments such as rain, snow, typhoon, dazzling and heavy fog, when normal driving is disturbed, if the problems that sight lines are fuzzy, visual field blind areas are encountered, vehicles are transversely unstable, steering wheels cannot be kept and the like occur, the vehicles can timely make corresponding reactions according to the simulation environment, drivers are prompted to be warned, and the vehicles are controlled to adopt active braking and active steering operation through a vehicle body stabilizing system according to real-time vehicle conditions.

Obstacle collision avoidance protection function: the method comprises the steps of utilizing visual simulation rendering to simulate collision of pedestrians and vehicles, designing a control strategy and a proper optimization algorithm, carrying out steering wheel shake reminding and safety prompting warning on the vehicles according to dangerous conditions based on motion prediction and risk assessment of the pedestrians and the vehicles, designing danger levels, and after triggering danger level signals, taking actions such as prompting warning, steering wheel shake, active steering, active braking and emergency braking in sequence to guarantee driving safety.

The invention integrates interactive, dynamic visual and physical behaviors, applies virtual reality technology to intelligent driving safety protection research, and realizes computer control, data real-time monitoring and visualization; different safety protection problem test scenes including electronic equipment faults, environmental interference, misoperation of a driver, driving fatigue, dangerous collision and other scenes are reproduced, a safety protection control strategy and an optimization algorithm are provided, and intelligent driving active safety and driving assistance intellectualization are improved; carrying out subjective and objective evaluation and verification on algorithm effectiveness and actual efficacy of the system on related technologies by using an intelligent driving safety protection system; the hardware platform of the intelligent driving safety protection system can restore the driving state of a real vehicle with high precision and high efficiency by combining software and hardware, and can combine dangerous scenes under virtual reality to test and verify safety protection strategies by taking a driver into consideration, so that the safety in development of the intelligent driving vehicle is greatly improved, and the intelligent driving safety protection system is a set of test system with complete functions.

Aiming at the safety protection problem in typical intelligent driving, the safety dangerous scene in intelligent driving is reproduced through the simulation scene of a visual simulation system and a designed hardware platform, an effective and safe protection strategy is provided, after dangerous conditions such as electronic equipment faults, environmental interference, misoperation of a driver, driving fatigue and dangerous collision occur, safety measures can be taken timely, an active safety signal is sent to a driving simulation test platform through a computer control module, a control instruction is sent to equipment such as a real braking system, a steering wheel, a seat and the like, and the real vehicle state change is fed back to the computer control system under the condition of the vehicle, so that closed-loop feedback control is formed, the vehicle is enabled to reach a safe state, and the provided safety protection strategy is tested and verified.

The system provided by the invention adopts a software and hardware combined mode to perform hardware-in-loop test from the perspective of safety protection, so that the state information of the vehicle is really monitored, and the safety protection strategy is applied to the real vehicle, so that dangerous scenes can be reproduced by depending on a test platform, the efficiency of testing and verifying the safety protection strategy is improved, and the safety of intelligent driving is greatly improved.

It should be noted that the above-mentioned embodiments enable a person skilled in the art to more fully understand the invention, without restricting it in any way. Therefore, although the present invention has been described in detail with reference to the drawings and examples, it will be understood by those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention.

Claims (10)

1. An intelligent driving safety protection system based on virtual reality is characterized by comprising a software simulation module, a hardware design and platform building module, a computer control module and a safety protection control strategy module, wherein the software simulation module is connected with the hardware design and platform building module;

the software simulation module comprises a vehicle dynamics modeling unit, a terrain modeling and visual simulation rendering unit and a driving simulation interaction unit, wherein the vehicle dynamics modeling unit analyzes a vehicle dynamics system according to an object-oriented theory, sets the matching data of the performance parameters of the whole vehicle, establishes a vehicle dynamics model based on virtual reality and realizes the motion simulation of the vehicle in a virtual prototype; the terrain modeling and visual simulation rendering unit establishes a three-dimensional high-definition terrain scene model based on a three-dimensional modeling technology aiming at safety threats in intelligent driving, and processes the three-dimensional high-definition terrain scene model through a real-time simulation program, an image generator and a visual effect database so as to realize virtual driving visual simulation rendering of the automobile; the driving simulation interaction unit utilizes a virtual reality technology to design an automobile driving virtual simulation experiment, and carries out automobile driving dynamic simulation, collision detection and multi-view switching;

the hardware design and platform building module comprises a hardware test platform design unit and a real-time data acquisition unit, wherein the hardware test platform design unit is used for designing a multifunctional multi-degree-of-freedom driving simulation test platform with human-vehicle-environment interaction characteristics, simulating a moving part and an operating mechanism of a cockpit, and realizing simulation of realistic tactile feeling in the driving process of a real vehicle in the cockpit; the real-time data acquisition unit is used for acquiring the operation data of a driver in real time so as to judge the state of the driver and the current state of the vehicle and realize effective real-time communication between the output equipment and the simulation;

the computer control module receives vehicle attitude signals transmitted from the terrain modeling and visual simulation rendering unit, and the vehicle attitude signals are converted into corresponding analog quantity control driving simulation test platforms through calculation, so that the man-machine interaction simulation attitude in the cockpit is realized, safety protection control strategies and algorithms are verified and optimized, and the danger early warning threshold value and the driving strategy set by the vehicle are adaptively adjusted;

the safety protection control strategy module makes a safety protection control strategy and an optimization algorithm aiming at safety threats existing in intelligent driving, and a vehicle is designed to adopt corresponding safety operation in a corresponding scene, so that active safety and auxiliary driving of intelligent driving are realized.

2. The intelligent driving safety protection system based on virtual reality according to claim 1, wherein the vehicle dynamics modeling unit adopts a simulation platform and virtual simulation software to establish a vehicle dynamics model based on virtual reality, the simulation platform adopts a Simulink simulation platform, and the virtual simulation software adopts Carsim multi-body dynamics simulation software.

3. The virtual reality-based intelligent driving safety protection system according to claim 1, wherein the safety threats in the terrain modeling and vision simulation rendering unit include at least one of equipment failure, system function limitations, personnel misoperation, dangerous collision and special environmental disturbance.

4. The intelligent driving safety protection system based on virtual reality as claimed in one of claims 1 to 3, wherein the driving simulation test platform in the hardware test platform design unit comprises a steering wheel, a throttle, a brake, a dynamic seat, an induction helmet and a simulation display, and has three-dimensional display, position capture and head tracking functions.

5. The virtual reality based intelligent driving safety protection system according to claim 4, wherein the manipulation data in the real-time data acquisition unit comprises at least one of opening degree of a throttle, steering wheel rotation angle, brake pedal angle and gear lever position.

6. The virtual reality-based intelligent driving safety protection system according to claim 4, wherein the hardware design and platform building module further comprises a hardware platform function verification unit, and the hardware platform function verification unit is combined with the real-time data acquisition unit and used for verifying the consistency, consistency and accuracy of the driving simulation test platform and the automobile virtual driving visual simulation rendering.

7. The virtual reality-based intelligent driving safety protection system according to one of claims 1 to 3, wherein the computer control module further performs I/O channel design, interface circuit design, anti-interference capability design and human-computer interaction interface design.

8. The virtual reality-based intelligent driving safety protection system according to claim 4, wherein the safety protection control strategy module comprises at least one of a designated driver state monitoring strategy, a driver misoperation protection strategy, a severe environment protection strategy and an obstacle collision avoidance protection strategy; when the driver misoperation or the operation exceeding the danger threshold value is detected, the vehicle is informed to give a prompt warning, and the vehicle is controlled to adopt active braking and active steering operation according to the real-time state; when the severe environment interferes with normal driving, the vehicle is informed to respond and prompt warning in time, and the vehicle is controlled to adopt active braking and active steering operation according to the real-time state; and designing a control algorithm when vehicle collision is simulated, and informing the vehicle to safely avoid collision according to dangerous conditions.

9. The virtual reality-based intelligent driving safety protection system according to claim 3, wherein the terrain modeling and view simulation rendering unit adopts a Multigen Creator to build a three-dimensional high-definition terrain scene model.

10. The virtual reality-based intelligent driving safety protection system according to claim 7, wherein the computer control module adopts a PC bus structure.

Images