Driving simulation system and method supporting manned and unmanned mixed driving test

Abstract

The invention discloses a driving simulation system and method supporting a manned and unmanned mixed driving test, and relates to the field of driving simulation. The cockpit unit comprises a plurality of steering wheel suites; for any steering wheel suite, the processor unit collects operation instruction information of a driver and determines a driving mode; when the driving mode is manned driving, calculating the pose information of the vehicle at the next moment by adopting a vehicle dynamics model based on the control instruction of the driver; when the driving mode is unmanned driving, calculating the pose information of the vehicle at the next moment according to the control instruction based on an automatic driving algorithm, reversely deducing a vehicle motion deflection angle and a steering wheel rotation angle to generate a control instruction, and feeding back the control instruction to the steering wheel suite; and performing instruction response according to the control instruction. Through the highly integrated and intelligent design, the problems that the application scene is single, algorithm control is not convenient enough, a feedback mechanism is insufficient, and a takeover detection mechanism is imperfect are solved.

Description

Technical Field

[0001] The present application relates to the field of driving simulation, and in particular to a driving simulation system and method that supports mixed manned and unmanned traffic testing. Background Art

[0002] As an important training and testing tool, driving simulators have been widely used in the automotive industry, traffic safety research, and driver training in recent years. With the rapid development of automotive technology, especially the rise of autonomous driving, the demand for driving simulators is growing. They not only provide a safe training environment for drivers but also offer an efficient data collection and analysis platform for automakers and research institutions.

[0003] Traditional driving simulators rely primarily on mechanical structures and simple visual feedback systems. While these systems can meet basic driving training requirements, they lack significant realism, interactivity, and data accuracy. With advances in virtual reality (VR), augmented reality (AR), and artificial intelligence (AI), modern driving simulators are increasingly becoming more realistic, intelligent, and multifunctional. For example, through high-resolution displays and multi-degree-of-freedom motion platforms, simulators can more realistically reproduce the visual, auditory, and physical feedback experienced during driving. Furthermore, AI-based driving behavior analysis and scenario generation technologies enable simulators to dynamically adjust training difficulty and test scenarios to better suit the needs of diverse users. However, these simulators still suffer from several drawbacks.

[0004] 1. Single application scenario: Most simulators only support input from a single device and have difficulty supporting the collaborative work of multiple users or multiple devices. They lack the ability to simulate mixed traffic scenarios, which limits research on mixed traffic scenarios and makes it difficult to meet the testing requirements of intelligent driving algorithms in mixed traffic environments.

[0005] 2. Algorithm control is not convenient enough: Existing simulators usually rely on external display interfaces or external devices to control the closed state of the algorithms installed in the simulator. However, the algorithms installed in this simulator can be independently turned on / off controlled by the steering wheel button algorithm unit, which makes the control of the algorithm more convenient and simple.

[0006] 3. Insufficient feedback mechanism: Existing simulators lack steering wheel angle feedback and automatic steering wheel return mechanisms in unmanned driving states. They are unable to simulate the dynamic response of the steering wheel in autonomous driving mode in real life, nor the self-righting of the steering wheel in human-driven state when the driver does not operate the steering wheel. This is disconnected from the actual vehicle motion state, reducing the realism of the simulation.

[0007] 4. Imperfect takeover detection mechanism: Most simulators only use a single driver takeover detection mechanism, which cannot cope with the complex behaviors that drivers may take when encountering risks and cannot effectively take over. Summary of the Invention

[0008] The purpose of this application is to provide a driving simulation system and method that supports mixed manned and unmanned traffic testing, which can solve the pain points of single application scenarios, inconvenient algorithm control, insufficient feedback mechanism, and imperfect takeover detection mechanism through highly integrated and intelligent design.

[0009] To achieve the above objectives, this application provides the following solutions:

[0010] In a first aspect, the present application provides a driving simulation system supporting manned and unmanned mixed traffic testing, comprising: a cockpit unit and a processor unit;

[0011] Wherein, the cockpit unit includes a plurality of steering wheel kits; the plurality of steering wheel kits are all connected to the processor unit;

[0012] For any of the steering wheel kits, the processor unit is configured to:

[0013] collecting the driver's operation command information based on the steering wheel kit;

[0014] Determining a driving mode based on the operation instruction information; the driving mode being manned driving or unmanned driving;

[0015] When the driving mode is manned, the vehicle's position information at the next moment is determined using a vehicle dynamics model based on the driver's control instructions; the position information includes vehicle posture information and position information; the vehicle dynamics model is a physical model determined by building a suspension system, steering system, tire model, braking system, and powertrain system based on dynamic parameters;

[0016] When the driving mode is unmanned driving, an automatic driving algorithm is used to determine the vehicle's posture information at the next moment according to the driver's control instructions, and to infer the vehicle's motion deflection angle and steering wheel angle to generate control instructions to be fed back to the steering wheel kit, and based on the detection result corresponding to the takeover detection mechanism, a command response is performed according to the control instruction; the detection result is that the driver takes over to exit the unmanned driving mode or continue unmanned driving.

[0017] In a second aspect, the present application provides a driving simulation method that supports manned and unmanned mixed traffic testing, wherein the driving simulation method that supports manned and unmanned mixed traffic testing is implemented using a driving simulation system that supports manned and unmanned mixed traffic testing; the method comprises:

[0018] Collect the driver's operating command information based on the steering wheel kit;

[0019] Determining a driving mode based on the operation instruction information; the driving mode being manned driving or unmanned driving;

[0020] When the driving mode is manned, the vehicle's position information at the next moment is determined using a vehicle dynamics model based on the driver's control instructions; the position information includes vehicle posture information and position information; the vehicle dynamics model is a physical model determined by building a suspension system, steering system, tire model, braking system, and powertrain system based on dynamic parameters;

[0021] When the driving mode is unmanned driving, an automatic driving algorithm is used to determine the vehicle's posture information at the next moment according to the driver's control instructions, and to infer the vehicle's motion deflection angle and steering wheel angle to generate control instructions to be fed back to the steering wheel kit, and based on the detection result corresponding to the takeover detection mechanism, a command response is performed according to the control instruction; the detection result is that the driver takes over to exit the unmanned driving mode or continue unmanned driving.

[0022] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0023] The present application provides a driving simulation system and method that supports mixed driving tests for manned and unmanned vehicles. The cockpit unit includes multiple steering wheel kits. For any steering wheel kit, the processor unit collects the driver's operating instruction information and determines the driving mode. When the driving mode is manned, the vehicle dynamics model is used to determine the vehicle's posture information at the next moment based on the driver's control instructions. When the driving mode is unmanned, the automatic driving algorithm is used to determine the vehicle's posture information at the next moment based on the driver's control instructions, and the vehicle's motion deflection angle and steering wheel angle are reversed to generate control instructions to be fed back to the steering wheel kit. Based on the detection results corresponding to the takeover detection mechanism, a command response is performed according to the control instruction to output the control instruction according to the vehicle's motion state and feed it back to the steering wheel kit to form a positive feedback force. In addition, detection is performed based on the takeover detection mechanism to solve the problem of imperfect takeover detection mechanism in the prior art. Based on different algorithm controls corresponding to different driving modes, the pain points of single application scenarios and inconvenient algorithm control are solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 The structural diagram of the driving simulation system to support mixed manned and unmanned traffic testing;

[0026] Figure 2 This is the hardware structure diagram of the driving simulation system;

[0027] Figure 3 This is a schematic diagram of the hardware connection method of the driving simulation system;

[0028] Figure 4 This is the software structure diagram of the driving simulation system;

[0029] Figure 5 This is an integrated diagram of the steering wheel's multi-function buttons;

[0030] Figure 6 This is a framework diagram of manned / unmanned mixed traffic based on multiple steering wheel kits;

[0031] Figure 7 This is a schematic diagram of mixed traffic with people and no one;

[0032] Figure 8 Flowchart of the driving simulation method to support mixed manned and unmanned traffic testing. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0034] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0035] In an exemplary embodiment, Figure 1 As shown, a driving simulation system supporting manned and unmanned mixed traffic testing is provided, including: a cockpit unit and a processor unit.

[0036] The cockpit unit includes a plurality of steering wheel kits; and the plurality of steering wheel kits are all connected to the processor unit.

[0037] For either steering wheel kit, the processor unit is used to:

[0038] The steering wheel kit collects the driver's operating command information; the driving mode is determined based on the operating command information; the driving mode is manned or unmanned; when the driving mode is manned, the vehicle's position information at the next moment is determined based on the driver's control command using the vehicle dynamics model; the position information includes: vehicle posture information and position information; the vehicle dynamics model is a physical model determined by building the suspension system, steering system, tire model, braking system and powertrain system based on dynamic parameters. When the driving mode is unmanned, the automatic driving algorithm is used to determine the vehicle's position information at the next moment based on the driver's control command, and the vehicle's motion deflection angle and steering wheel angle are inferred to generate control commands to be fed back to the steering wheel kit, and based on the detection results corresponding to the takeover detection mechanism, the command response is carried out according to the control command; the detection result is that the driver takes over to exit the unmanned driving mode or continue unmanned driving.

[0039] In one embodiment, the driving simulation system supporting manned and unmanned mixed traffic testing further includes: an information display unit.

[0040] The processor unit is connected to the information display unit; the processor unit is used to generate a visual signal based on the posture information of the next moment corresponding to manned driving and / or the posture information of the next moment corresponding to unmanned driving, combined with the simulation scene; the simulation scene is obtained by using VTD software to perform vehicle scene simulation and visual modeling processing.

[0041] The information display unit is used to determine a video signal according to the visual signal for visual display.

[0042] As an optional embodiment, the cockpit unit further includes: a base, a cockpit body and a control console.

[0043] The cockpit body is fixed to the base by welding, and the front and rear movable end slides of the cockpit body are fixed to the base by screws; the steering wheel kit is fixed to the extension bracket of the base by a knob.

[0044] The control panel is a detachable design structure, and is fixed and separated from the base by screws.

[0045] The steering wheel kit is connected to the processor unit via a USB interface.

[0046] The information display unit includes a plurality of high-resolution displays; a high-resolution display is a display with a screen resolution higher than a set threshold.

[0047] The driving simulation system mentioned in this application is highly integrated and multifunctional. The highly integrated feature is mainly reflected in the hardware and software levels of the system, and the multifunctional feature is mainly reflected in the practical application level.

[0048] like Figure 2 and Figure 3 As shown in the figure, at the hardware level, the system is integrated into three parts: the cockpit module (cockpit unit), the information display module (information display unit), and the processor module (processor unit). These are responsible for collecting driver information, calculating control commands, and visually presenting dynamic information, respectively. The specific module components are as follows.

[0049] ① The cockpit module, i.e. the cockpit unit, consists of a base, a cockpit body, a steering wheel kit (the G29 steering wheel kit (steering wheel + pedals) can be selected), and a control panel. In the cockpit module itself, the cockpit body is fixed to the base by welding, the front and rear seat sliding slots are fixed to the base by screws, the steering wheel kit is fixed to the base extension bracket by knobs, and the control panel is detachable, with screws used to fix and separate the console and the base. The seven displays in the information display module, the 57-inch curved display 1 and the 13.3-inch rectangular displays 2-5 are fixed to the bracket in front of the base by screw fasteners and the bottom slots. The two servers in the processor module are placed inside the control panel.

[0050] The cockpit module is mainly used to collect driver operation command information and transmit the relevant command information to the server module through SCP messages that can be recognized by the ROS system. The base serves as the simulator support platform, providing stable support, and internally inherits power cords, device connection cables, interior wall table lamps, power supplies and other equipment; the cockpit is designed with reference to real vehicle seats to ensure that the driver can get a more realistic sense of support; the steering wheel kit can simulate the steering wheel rotation of the vehicle in a virtual environment, and supports the adjustment of the rotation angle and sensitivity. The brake and accelerator pedals can simulate the acceleration and braking effects of a real vehicle. The steering wheel is also equipped with a force feedback function, which can simulate the dynamic response in real driving when the automatic driving mode is turned on.

[0051] ② Information display module: It consists of seven high-resolution displays, which respectively display the driver's main perspective, left and right rearview mirrors, vehicle instrument panel, running map, operation interface and central control interface. Among them, the 57-inch curved display 1 displays the main perspective; the 13.3-inch rectangular display 2 displays the left rearview mirror; the 13.3-inch rectangular display 3 displays the right rearview mirror; the 13.3-inch rectangular display 4 displays the user interface; the 13.3-inch rectangular display 5 displays the three-dimensional map; the 27-inch rectangular display 6 and display 7 display the central control interface, providing the driver with a full range of road traffic flow-related feedback information.

[0052] ③ Processor Module: Consists of two high-performance servers. Server 1 is primarily responsible for running VTD and is also responsible for receiving and processing input signals from multiple steering wheel kits, ensuring real-time transmission of operating commands. It also visualizes vehicle position information (such as position, speed, and direction) through the VTD module and provides feedback to the driver, enhancing the immersiveness and interactivity of the simulation. It also supports multi-display output, providing the driver with comprehensive visual feedback, including the main view, rearview mirror view, 3D map, and user interface. Server 2 is used to run Trucksim and Matlab / Simulink, and is responsible for providing the vehicle's dynamic model and autonomous driving algorithms. The two servers serve as intermediate modules for software and hardware interaction, primarily responsible for receiving driver input, converting control commands, and providing central control interface data.

[0053] like Figure 4 As shown in the figure, at the software level, the driving simulation system can establish a joint simulation platform based on VTD-Trucksim-Matlab, where:

[0054] ①VTD software: mainly responsible for the appearance design of vehicle models, dynamic simulation scene design and static road design.

[0055] In the vehicle model appearance design, VTD achieves high-precision visual modeling of multi-axle vehicles, including details such as body rendering, wheel steering control, and headlight on and off effects.

[0056] In the design of dynamic simulation scenarios, VTD supports the simulation of dynamic events such as random traffic flow, interfering vehicle control, pedestrian obstacle addition, weather changes (such as rain, snow, fog, etc.) and day and night alternation.

[0057] In static road design, VTD provides a variety of test scenarios, such as curve scenarios for LKA (lane keeping assist) testing, straight road scenarios for AEB (automatic emergency braking) testing, parking scenarios for PA (parking assist) testing, and following scenarios for ACC (adaptive cruise control) testing, ultimately presenting the vehicle's motion status in a visual form.

[0058] ②Trucksim software: responsible for the construction and calculation of the vehicle dynamics model. First, define the vehicle type in Trucksim. The model built in this application is a multi-axle vehicle model. Enter the key dynamic parameters of the multi-axle vehicle in Trucksim: such as vehicle mass, number of axles and their positions, wheelbase (the distance from each axle to the center of mass), vehicle position, center of mass position, and moment of inertia. Then, build the four subsystems of the multi-axle vehicle: suspension system, steering system, tire model, braking system, and powertrain system. In the suspension system, define the model suspension type, set the spring stiffness curve, configure the shock absorber characteristics, and define the suspension limit block parameters; in the steering system, set the steering mechanism geometry, define the steering transmission ratio, and configure the power assist characteristics; in the tire model, select the tire model, enter the tire characteristic parameters, and set the tire-road friction coefficient; in the braking system, configure the brake type and parameters, and set the brake distribution ratio; in the powertrain system, select the engine type, enter the engine external characteristic curve, and set the drive axle speed ratio to generate the corresponding complete dynamic model of the multi-axle vehicle.

[0059] Trucksim receives control commands from Simulink and calculates the vehicle's state information (such as position, velocity, acceleration, etc.) at the next moment based on the dynamic model. ① Vehicle posture information acquisition: When Trucksim receives the driving signal input from the steering wheel, the signal passes through the steering system and calculates the steering wheel angle based on the set steering gear ratio and steering system stiffness. The tire system calculates the lateral and longitudinal slip rates of each tire based on the current received vehicle speed signal, wheel speed, and steering angle, ultimately obtaining the longitudinal force, lateral force, self-aligning torque, and vertical force of each tire. The suspension system calculates the suspension motion trajectory and changes in wheel alignment parameters based on the hard point coordinates to determine whether the vehicle is in a roll condition. ② Vehicle position information acquisition: The current vehicle speed signal, wheel speed, and steering angle are input into Trucksim, and then the vehicle dynamics in the model are solved to obtain acceleration. The acceleration is kinematically integrated to obtain the new velocity and position, which are then updated to all subsystem states. The calculated results are fed back to Simulink for subsequent simulation and visualization.

[0060] ③MATLAB / Simulink software: This software takes on the key tasks of signal transmission and algorithm implementation. Simulink establishes the communication protocol between the hardware (steering wheel kit) and the three software components, converting hardware signals into a format recognizable by the co-simulation software and supporting data transmission between Ubuntu and Windows systems. Simulink also integrates four autonomous driving algorithms: ACC (adaptive cruise control), AEB (automatic emergency braking), LKA (lane keeping assist), and PA (parking assist), enabling the co-simulated vehicle to possess adaptive cruise control, automatic emergency braking, lane keeping, and automatic parking capabilities. In autonomous driving mode, Simulink infers the tire deflection angle required for vehicle movement based on the vehicle's next position information provided by the autonomous driving algorithm. The steering wheel angle required for actual driving is calculated based on the proportional relationship between the tire and the steering wheel. This is converted into control commands via the communication module in Simulink, and the signals are transmitted to the steering wheel kit's force feedback motor, enabling steering wheel force feedback in autonomous driving mode.

[0061] like Figure 5 As shown, at the practical application level, the system has the integrated implementation of steering wheel multi-function buttons, dynamic response implementation of the steering wheel, takeover design in automatic driving mode, and implementation of manned / unmanned mixed driving simulation scenario functions based on multiple steering wheel kits.

[0062] ① Integration of multi-function buttons: The driver can switch between manned and unmanned driving modes through the Enter button of the steering wheel kit, and control the opening and closing of ACC, AEB, PA, and LKA algorithms through specific buttons on the steering wheel (upper triangle, circle, cross, and square buttons), and use the left / right paddles to activate the left / right lane change algorithm. When the relevant button is pressed, the steering wheel kit will convert the driver's operation signal into a digital signal, and send the relevant Topic to the receiving module in Simulink through Ros: When the Enter button is pressed, the digital signal in the receiving module will be converted from 0 to 1, and the signal will be sent to the judgment module of the main switch of the autonomous driving algorithm under the Simulink module. The module will judge that every time there is a signal transition (0→1), the state of the Switch switch will change. When the main switch is in the off state, the output value of the autonomous driving state is 0, and the signals generated by the upper triangle, circle, cross, square buttons and paddles will not be sent to Corresponding algorithm modules; When the main switch is on, the automatic driving state output value is 1. Pressing any of the upper triangle, circle, cross, square, or paddle buttons will cause the digital signal module corresponding to the button to switch from 0 to 1. The signal will be sent to the judgment module of the corresponding algorithm module, among which the upper triangle signal is sent to the ACC algorithm module, the circle signal is sent to the AEB algorithm, the cross signal is sent to the PA algorithm module, the square signal is sent to the LKA algorithm, and the paddle signal is sent to the lane change algorithm. According to the judgment, each time the signal switches from 0 to 1, the switch state of the corresponding algorithm module will be switched. When the main switch is on and the Enter signal changes from 0 to 1, the module will forcibly switch the Switch of all algorithms to the off state.

[0063] Implementation of the dynamic response of the steering wheel: In order to simulate the dynamic response of the steering wheel in the unmanned driving state, the assisted driving algorithm will calculate the vehicle's motion state at the next moment based on the previous moment's posture information transmitted by VTD and the dynamic model in Trucksim, and output the corresponding posture information to Simulink. At the same time, Simulink will reversely infer the deflection angle required for vehicle movement based on the vehicle's next moment posture information given by the autonomous driving algorithm, and calculate the steering wheel angle required for actual driving through the proportional relationship between the tire and the steering wheel. The communication module in Simulink converts it into a control command to output the required steering wheel angle command, and applies the corresponding torque to the steering wheel through the force feedback motor to reproduce the dynamic response of the autonomous driving steering wheel under real conditions; under human driving conditions, the steering wheel will self-center when the driver does not operate the steering wheel.

[0064] ③ Takeover design in automatic driving mode: Two takeover detection mechanisms are integrated in the Simulink module of the simulator system. The detection mechanism is implemented through the steering wheel kit signal access part and the automatic driving algorithm part in the Simulink module. The automatic driving algorithm part detects whether the state output value of the current algorithm is 1 (that is, whether it is in an activated state). Then, it judges whether the driver takes over based on the signal given by the steering wheel kit. The two takeover detection mechanisms are steering wheel angle detection and brake pedal detection. The specific implementation process is as follows: When the simulated vehicle is in automatic driving mode (when the state output value of the algorithm is 1), the steering wheel signal and the brake pedal signal will be transmitted to the takeover judgment module in the algorithm module. This module is divided into two judgment logics, which controls the main switch of the automatic driving algorithm by judging "whether the driver turns the steering wheel" or "whether the driver steps on the brake pedal". Status value: ① Determine whether the driver turns the steering wheel: When in the automatic driving state, the steering wheel transmits the driving signal of the driver manipulating the steering wheel and compares it with the steering wheel angle calculated by the algorithm. When the error between the two exceeds 30° (adjustable), it is considered that "the driver turns the steering wheel". At this time, the status value of the main switch is changed to 0, and the vehicle exits the automatic driving; ② Determine whether the driver steps on the brake pedal: When the vehicle is in the automatic driving state, the pedal signal in the steering wheel kit (the initial signal is -1, and the signal range is -1 to 1) will be input to the automatic driving takeover judgment module. The current pedal signal output value will be compared with the pedal signal output value at the previous moment. If there is no difference between the front and rear signals, it is considered that the driver did not step on the pedal. When the system detects that the steering wheel angle is greater than the steering wheel angle calculated by the algorithm, or when it detects that the driver lightly steps on the brake pedal, the system will automatically exit the unmanned driving mode.

[0065] ④ Realization of manned / unmanned mixed driving simulation scene functions based on multiple steering wheel kits, such as Figure 6 As shown, the system supports the integration of steering wheels from multiple steering wheel kits. The steering wheel kits are connected to the computer via USB. By connecting multiple steering wheel kits to different USB ports, configuring a corresponding device ID for each port, and distinguishing different steering wheel kits by identifying different device IDs in the software, each device ID is configured in the Simulink communication protocol with the port number for sending and receiving VTD SCP message data and Ubuntu system Topic data during communication. This allows different steering wheel kits to control different vehicles through their corresponding device IDs. The simulation platform can independently control each vehicle and dynamically assign operating modes—some vehicles are controlled using preset autonomous driving algorithms, while others are operated in real time by the driver using human-computer interaction devices, thus enabling mixed manned / unmanned driving scenarios.

[0066] The advantages of this application are:

[0067] 1. Highly integrated and manned / unmanned mixed traffic simulation scenario construction.

[0068] The simulation system is not only highly integrated and scalable at the software and hardware levels, but also supports the access of steering wheels from multiple steering wheel kits. In the same simulation scenario, the system can achieve independent control of different vehicles by accessing the steering wheel devices of multiple steering wheel kits. In the simulation, the vehicle operation mode can be flexibly configured: some vehicles are controlled by preset autonomous driving algorithms, while other vehicles are operated in real time by the driver through the steering wheel, thus creating a mixed manned / unmanned driving simulation environment, simulating the coexistence of manned and unmanned vehicles in real traffic environments. Through the collaborative work of multiple steering wheels, the system can meet the testing requirements of mixed manned / unmanned traffic scenarios.

[0069] 2. Highly realistic driving simulation and feedback mechanism.

[0070] The system significantly enhances the realism of the simulation through an information display module comprised of seven displays, a steering wheel angle feedback mechanism, and a takeover detection mechanism. The information display module displays the driver's primary perspective, left and right rearview mirrors, the vehicle instrument panel, a running map, an operating interface, and a central control interface, enhancing the driver's interactive experience during simulated driving. In autonomous driving, to simulate the dynamic response of the steering wheel, the assisted driving algorithm outputs steering wheel angle commands based on the vehicle's dynamic state and applies corresponding torque to the steering wheel via a force feedback motor to replicate the dynamic response of an autonomous driving steering wheel under real-world conditions. In human-driven driving, the steering wheel self-centers when the driver is not operating the steering wheel, enhancing the realism of the simulated driving. The combined use of two takeover detection mechanisms (steering wheel angle detection and brake pedal detection) enhances the safety of the simulated driving.

[0071] 3. Independent on / off control of algorithm unit.

[0072] Through the hardware interface of the steering wheel in the steering wheel kit, the driver can quickly switch driving modes and algorithm status. The driver switches between manned / unmanned driving modes through the Enter button on the steering wheel, and controls the opening and closing of ACC, AEB, PA, and LKA algorithms through specific buttons on the steering wheel (upper triangle, circle, cross, and square buttons), and uses the left / right paddles to activate the left / right lane change algorithm.

[0073] To sum up, the system mentioned in this application is superior to existing simulators in terms of high integration and construction of mixed manned / unmanned simulation scenarios, high-realism driving simulation and feedback mechanism, and independent on / off control of algorithm units. It can provide drivers with a more comprehensive and realistic intelligent driving training experience, and has broad application prospects and market value.

[0074] Specifically, the benefits of this application are:

[0075] Hardware integration, high-realism driving simulation, and independent on / off control of algorithm units.

[0076] (1) Multi-screen real-time interaction: The system is equipped with seven displays as information display modules, including the main view display screen, left and right rearview mirror display screens, user interface display screen, three-dimensional map display screen and central control interface. The main view display screen and left and right rearview mirrors provide the driver with a simulation scene that is as realistic as possible; the three-dimensional map display screen will display the trajectory planned by the algorithm in real time, so that the user can intuitively understand the vehicle's driving path; the user interface display screen will display the activation / disable status of the current algorithm, helping the driver to grasp the system operation status in real time; the central control interface can realize the selection of vehicle appearance model, vehicle dynamics model, static scene, and simulated dynamic scene, and can also control the activation and deactivation of the automatic driving algorithm in real time. The central control interface also has the function of real-time storage and export of simulation data.

[0077] (2) Steering wheel force feedback: In the unmanned driving state, the assisted driving algorithm will output the corresponding steering wheel angle to the steering wheel according to the vehicle's motion state, forming a positive feedback force, so that the steering wheel and the vehicle's motion state remain consistent, thereby improving the realism of the simulation.

[0078] (3) Two driver takeover mechanisms: The system supports two takeover detection mechanisms, including steering wheel angle detection and brake pedal detection. When the steering wheel angle is detected to be much greater than the steering wheel angle calculated by the algorithm, or when the brake pedal is detected to be lightly pressed, the system will automatically exit the autonomous driving mode to ensure driving safety.

[0079] (4) Independent on / off control of algorithm units: The Enter button on the steering wheel serves as the master switch for the assisted driving mode, allowing the driver to quickly switch between manned and unmanned driving modes. In the assisted driving mode, specific buttons on the steering wheel are used to control the individual on / off of the five intelligent driving algorithms: the upper arrow button is the switch for the ACC algorithm, the circle button is the switch for the AEB function, the upper triangle button is the switch for the PA algorithm, and the square button is the switch for the LKA algorithm. The left and right paddles are used to activate the lane change algorithm. Pressing the left paddle activates the left lane change algorithm, and pressing the right paddle activates the right lane change algorithm. When the ACC algorithm is activated, the driver can adjust the set vehicle speed using the plus and minus buttons on the steering wheel. The plus button increases the set vehicle speed, and the minus button decreases the set vehicle speed. The increase and decrease values ​​can be customized according to needs.

[0080] 2. Multiple simulators: Collaborative use of manned / unmanned mixed simulation scenarios, such as Figure 7 and Figure 8 shown.

[0081] (1) Construction of mixed manned and unmanned driving scenarios: The innovation of multiple simulators is mainly reflected in their support for multiple steering wheel access and the ability to realize mixed manned / unmanned driving simulation scenarios. In the same simulation scenario, the system can achieve independent control of different vehicles by accessing multiple steering wheel devices. In the simulation, the vehicle operation mode can be flexibly configured: some vehicles are controlled by preset autonomous driving algorithms, while other vehicles are controlled by the driver in real time through the steering wheel, thereby constructing a mixed manned / unmanned driving simulation environment and simulating the coexistence of manned and unmanned vehicles in a real traffic environment. This function not only provides a more realistic experimental scenario for the testing and optimization of intelligent driving algorithms, but also provides important technical support for research fields such as multi-vehicle cooperative driving and mixed manned / unmanned traffic flow simulation. The design of multiple steering wheel access further enhances the scalability and flexibility of the system, enabling it to adapt to the testing needs of more complex scenarios.

[0082] In an exemplary embodiment, a driving simulation method supporting a manned and unmanned mixed traffic test is provided, wherein the driving simulation method supporting a manned and unmanned mixed traffic test is implemented using a driving simulation system supporting a manned and unmanned mixed traffic test. Figure 8 As shown, the method includes:

[0083] Step 100: Collect the driver's operation instruction information based on the steering wheel kit.

[0084] Step 200: Determine a driving mode based on the operation instruction information. The driving mode can be manned driving or unmanned driving.

[0085] Step 300: When the driving mode is manned, the driver's control commands are used to determine the vehicle's next position information using the vehicle dynamics model. Position information includes vehicle posture information and position information. The vehicle dynamics model is a physical model determined by building models of the suspension system, steering system, tire system, braking system, and powertrain based on dynamic parameters.

[0086] Step 400: When the driving mode is unmanned, the autonomous driving algorithm determines the vehicle's next position based on the driver's control commands, and infers the vehicle's motion deflection angle and steering wheel angle to generate control commands that are fed back to the steering wheel assembly. Based on the detection results of the takeover detection mechanism, the control assembly responds to the control commands. The detection result indicates that the driver has taken over to exit the unmanned driving mode or continue unmanned driving.

[0087] In one embodiment, a method for determining a vehicle dynamics model specifically includes:

[0088] Obtain dynamic parameters; dynamic parameters include: vehicle mass, number and position of axles, wheelbase, vehicle position, center of mass position and moment of inertia.

[0089] Build the suspension system, steering system, tire model, braking system and power transmission system; in the suspension system, based on the suspension type, the corresponding spring stiffness curve is set, and the shock absorber characteristics and suspension limit block parameters are configured; in the steering system, the steering mechanism geometry is set, the steering transmission ratio is preset, and the power assistance characteristics are configured; in the tire model, the tire-road friction coefficient is set based on the tire characteristic parameters; in the braking system, the brake type and parameters are configured, and the braking distribution ratio is set; in the power transmission system, based on the engine type, the drive axle speed ratio is set according to the engine external characteristic curve.

[0090] The vehicle dynamics model is determined based on the dynamic parameters and the established suspension system, steering system, tire model, braking system and powertrain system.

[0091] The autonomous driving algorithm integrates four autonomous driving algorithms: adaptive cruise control, automatic emergency braking, lane keeping assist and parking assist.

[0092] The detection process corresponding to the takeover detection mechanism includes:

[0093] The driver's operation signal is collected based on the steering wheel kit; if the operation signal is a signal generated by steering wheel angle processing, then:

[0094] It is determined whether the steering wheel angle corresponding to the signal generated by the steering wheel angle processing is greater than the inversely derived steering wheel angle to obtain a first determination result.

[0095] If the first judgment result is yes, the detection result obtained is that the driver takes over and exits the unmanned driving mode.

[0096] If the first judgment result is no, then the error between the steering wheel angle and the inversely derived steering wheel angle is calculated, and it is determined whether the error is greater than a set threshold to obtain a second judgment result.

[0097] If the second judgment result is yes, the detection result obtained is that the driver takes over and exits the unmanned driving mode; if the second judgment result is no, the detection result obtained is that no driver takes over and unmanned driving continues.

[0098] If the operation signal is a signal generated by brake pedal processing, then:

[0099] Compare with the pedal signal output value at the previous moment to determine whether the comparison result is no difference, and obtain a third judgment result; if the third judgment result is yes, the detection result obtained is that there is no driver taking over and unmanned driving continues; if the third judgment result is no, the detection result obtained is that the driver takes over and exits the unmanned driving mode.

[0100] As an optional implementation, the method further includes:

[0101] Based on the next-moment pose information corresponding to a manned vehicle and / or the next-moment pose information corresponding to an unmanned vehicle, a visual signal is generated in conjunction with a simulated scene. The simulated scene is generated by using VTD software to simulate the vehicle scene and perform visual modeling. Based on the visual signal, a video signal is determined for visual display.

[0102] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0103] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the system, method, and core concept of this application. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of this application. In summary, the contents of this specification should not be construed as limiting this application.

Claims

1. A driving simulation system that supports mixed manned and unmanned traffic testing, characterized in that: include: Cockpit unit and processor unit; Wherein, the cockpit unit includes a plurality of steering wheel kits; the plurality of steering wheel kits are all connected to the processor unit; For any of the steering wheel kits, the processor unit is configured to: collecting the driver's operation command information based on the steering wheel kit; Determining a driving mode based on the operation instruction information; the driving mode being manned driving or unmanned driving; When the driving mode is manned, the vehicle's position information at the next moment is determined using a vehicle dynamics model based on the driver's control instructions; the position information includes vehicle posture information and position information; the vehicle dynamics model is a physical model determined by building a suspension system, steering system, tire model, braking system, and powertrain system based on dynamic parameters; When the driving mode is unmanned driving, an automatic driving algorithm is used to determine the vehicle's posture information at the next moment according to the driver's control instructions, and to infer the vehicle's motion deflection angle and steering wheel angle to generate control instructions to be fed back to the steering wheel kit, and based on the detection result corresponding to the takeover detection mechanism, a command response is performed according to the control instruction; the detection result is that the driver takes over to exit the unmanned driving mode or continue unmanned driving.

2. The driving simulation system supporting mixed manned and unmanned traffic testing according to claim 1 is characterized in that: The driving simulation system supporting manned and unmanned mixed traffic testing further includes: an information display unit; The processor unit is connected to the information display unit; The processor unit is used to generate a visual signal based on the posture information of the next moment corresponding to manned driving and / or the posture information of the next moment corresponding to unmanned driving, in combination with a simulation scene; the simulation scene is obtained by using VTD software to perform vehicle scene simulation and visual modeling processing; The information display unit is used to determine a video signal according to the visual signal for visual display.

3. The driving simulation system supporting mixed manned and unmanned traffic testing according to claim 1 is characterized in that: The cockpit unit further comprises: a base, a cockpit body and a control console; The cockpit body is fixed to the base by welding, and the front and rear movable end slides of the cockpit body are fixed to the base by screws; The steering wheel kit is fixed to the extension bracket of the base via a knob; The control panel is a detachable design structure, and the control panel is fixed to and separated from the base by screws.

4. The driving simulation system supporting mixed manned and unmanned traffic testing according to claim 1 is characterized in that: The steering wheel kit is connected to the processor unit via a USB interface.

5. The driving simulation system supporting mixed manned and unmanned traffic testing according to claim 2 is characterized in that: The information display unit includes a plurality of high-resolution displays; the high-resolution displays are displays with a screen resolution higher than a set threshold.

6. A driving simulation method supporting mixed manned and unmanned traffic testing, characterized in that: The driving simulation method supporting mixed manned and unmanned traffic testing is implemented by using the driving simulation system supporting mixed manned and unmanned traffic testing according to any one of claims 1 to 5; the method comprises: Collect the driver's operating command information based on the steering wheel kit; Determining a driving mode based on the operation instruction information; the driving mode being manned driving or unmanned driving; When the driving mode is manned, the vehicle's position information at the next moment is determined using a vehicle dynamics model based on the driver's control instructions; the position information includes vehicle posture information and position information; the vehicle dynamics model is a physical model determined by building a suspension system, steering system, tire model, braking system, and powertrain system based on dynamic parameters; When the driving mode is unmanned driving, an automatic driving algorithm is used to determine the vehicle's posture information at the next moment according to the driver's control instructions, and to infer the vehicle's motion deflection angle and steering wheel angle to generate control instructions to be fed back to the steering wheel kit, and based on the detection result corresponding to the takeover detection mechanism, a command response is performed according to the control instruction; the detection result is that the driver takes over to exit the unmanned driving mode or continue unmanned driving.

7. The driving simulation method supporting manned and unmanned mixed traffic testing according to claim 6, characterized in that: The method for determining the vehicle dynamics model specifically includes: Obtaining dynamic parameters; the dynamic parameters include: vehicle mass, number and position of axles, wheelbase, vehicle position, center of mass position, and moment of inertia; Build the suspension system, steering system, tire model, braking system, and powertrain system. In the suspension system, set the spring stiffness curve based on the suspension type, configure the shock absorber characteristics, and configure the suspension limiter parameters. In the steering system, set the steering mechanism geometry, preset the steering transmission ratio, and configure the power assist characteristics. In the tire model, set the tire-road friction coefficient based on the tire characteristic parameters. In the braking system, configure the brake type and parameters, and set the brake distribution ratio. In the powertrain system, set the drive axle speed ratio based on the engine type and the engine external characteristic curve. A vehicle dynamics model is determined based on the dynamic parameters and the constructed suspension system, steering system, tire model, braking system and power transmission system.

8. The driving simulation method supporting mixed manned and unmanned traffic testing according to claim 6, characterized in that: The autonomous driving algorithm integrates four autonomous driving algorithms: adaptive cruise control, automatic emergency braking, lane keeping assist and parking assist.

9. The driving simulation method supporting mixed manned and unmanned traffic testing according to claim 6, characterized in that: The detection process corresponding to the takeover detection mechanism specifically includes: Collect the driver's operation signals based on the steering wheel kit; If the operation signal is a signal generated by steering wheel angle processing, then: determining whether a steering wheel angle corresponding to a signal generated by performing steering wheel angle processing is greater than an inversely derived steering wheel angle, thereby obtaining a first determination result; If the first judgment result is yes, the detection result obtained is that the driver takes over and exits the unmanned driving mode; If the first judgment result is no, then calculating the error between the steering wheel angle and the inferred steering wheel angle, and determining whether the error is greater than a set threshold to obtain a second judgment result; If the second judgment result is yes, the detection result obtained is that the driver takes over and exits the unmanned driving mode; If the second judgment result is no, the detection result obtained is that no driver has taken over and the vehicle continues to be unmanned; If the operation signal is a signal generated by brake pedal processing, then: Comparing the pedal signal output value with the previous moment to determine whether the comparison result is no difference, thereby obtaining a third determination result; If the third judgment result is yes, the detection result obtained is that no driver has taken over and the vehicle continues to be unmanned; If the third judgment result is no, the detection result obtained is that the driver takes over and exits the unmanned driving mode.

10. The driving simulation method supporting manned and unmanned mixed traffic testing according to claim 6, characterized in that: Also includes: Generate a visual signal based on the next moment's posture information corresponding to the manned vehicle and / or the next moment's posture information corresponding to the unmanned vehicle in combination with the simulation scene; The simulation scene is obtained by using VTD software to perform vehicle scene simulation and visual modeling; A video signal is determined according to the visual signal for visual display.

Images