Audio and video driving scene simulation test system for expressway main line section traffic broadcast

Abstract

The application discloses a highway main line section traffic broadcast audio and video driving scene simulation test system.The system comprises an acoustic simulation subsystem, a visual simulation subsystem, a driving simulator interaction subsystem and a Python program integrated control subsystem; the acoustic simulation subsystem comprises a broadcast audio library, a background noise audio library and a volume adjustment module, realizes split track processing and dynamic volume adjustment of broadcast sound and background noise; the visual simulation subsystem comprises a full-scene visual simulation module and a display output module, the former constructs panoramic simulation visual content of a target test road section, and the latter presents a front field of view of a simulation scene and a driver's perspective picture of a real video; the driving simulator interaction subsystem, namely a driving operation peripheral module, controls a simulation vehicle based on a force feedback control signal of the Python program integrated control subsystem, and the Python program integrated control subsystem collects driving state data of the simulation vehicle and driver behavior data acquired by the driving operation peripheral module.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation systems and driving simulation testing technology, specifically to an audio-visual driving scenario simulation testing system for traffic broadcasting on the main line of highways, which is suitable for evaluating the driver's real feedback to the highway broadcasting system under different vehicle speeds and noise environments. Background Technology

[0002] With the continuous increase in highway mileage and the sustained growth in motor vehicle ownership, highway traffic flow is increasing daily. Highway broadcasting is a core channel for conveying important information such as road conditions, warnings, and emergency commands. Because of the complex noise environment and limited driver attention on highways, the actual playback effect of traffic broadcasts significantly depends on the playback strategy and volume. Therefore, it is necessary to investigate the impact of different playback strategies and volumes on driver information reception to determine the optimal playback strategy and volume for highway traffic broadcasting. However, conducting on-site tests of traffic broadcasting directly on highways is not only costly but also poses significant safety hazards. Simply collecting in-vehicle audio during driving cannot reflect the impact of driver distraction and therefore fails to demonstrate the true playback effect of traffic broadcasting on highways. Therefore, setting up a driving simulator environment in the laboratory to simulate and test highway traffic broadcasting not only realistically reflects the playback effect of traffic broadcasts but also reduces testing costs and safety hazards.

[0003] Existing research on highway traffic broadcast testing mainly focuses on signal transmission optimization, sound quality testing in ordinary scenarios, and improvements to traditional real-vehicle road tests. There is limited research on audio-visual driving scenario simulation testing in highway mainline scenarios. Patent CN120412301A utilizes LTE-V technology to receive traffic event information, performs noise reduction and enhancement on voice broadcasts, and includes a simplified sound quality testing module. However, this method does not involve video and simulation testing, and can only evaluate voice quality, making it difficult to meet the joint simulation requirements of complex mainline scenarios. Patent CN120766520B generates and broadcasts traffic diversion alert information, including a simplified testing scheme for information transmission effectiveness. However, this method does not consider the impact of dynamic traffic flow on broadcast effectiveness. Patent CN118405148A determines the optimal voice broadcast mode based on simulated driving behavior data, which can assist in optimizing voice broadcast effects. However, this method only includes voice broadcast mode testing, has a single testing dimension, and the verification conditions are limited to preset dangerous scenarios, limiting its application scope.

[0004] Against this backdrop, the common problems existing in the traffic broadcast testing methods described in the above patents include: failure to implement joint audio-visual simulation testing, failure to consider complex environments such as dynamic traffic flow, and a single testing dimension. Traditional field testing is costly, inefficient, and poses safety hazards. How to fully utilize simulation testing technology to accurately simulate various scenarios on the main line of highways, comprehensively evaluate the playback quality and coverage of traffic broadcast audio and video, thereby replacing field testing, reducing testing costs, and improving testing efficiency and accuracy, has become an urgent problem to be solved. Summary of the Invention

[0005] The technical problem to be solved by this invention is to provide a simulation testing system that can replace on-site testing, realize audio-video joint simulation, and comprehensively evaluate the effect of traffic broadcasting on the main line of highways, addressing the problems existing in the current highway traffic broadcasting testing methods, such as the lack of audio-video joint simulation, the failure to consider complex environments such as dynamic traffic flow, and the single testing dimension.

[0006] To address the aforementioned technical problems, this invention provides an audio-visual driving scenario simulation testing system for traffic broadcasting on the main line of a highway. This system includes an acoustic simulation subsystem, a visual simulation subsystem, a driving simulator interaction subsystem, and a Python program integrated control subsystem.

[0007] The acoustic simulation subsystem includes a broadcast audio library, a background noise audio library, and a volume adjustment module, which performs track-by-track processing and dynamic volume adjustment for broadcast sound and background noise; the visual simulation subsystem includes a full-scene visual simulation module and a display output module, wherein the full-scene visual simulation module is used to construct the full-scene simulation visual display content of the target test road section, and the display output module is used to display the driver's forward field of view in the simulation scene and the driver's perspective in the real video data.

[0008] The driving simulator interaction subsystem, namely the driving operation peripheral module, realizes the linkage control between the driving operation peripheral module and the simulated vehicle based on the force feedback control signal of the Python program integrated control subsystem. The Python program integrated control subsystem simultaneously collects the driving status data of the simulated vehicle (such as steering wheel angle, pedal opening, etc.) and the driver behavior data (speed, acceleration, position, vehicle internal parameters, etc.) obtained by the driving simulator interaction subsystem.

[0009] In the acoustic simulation subsystem, the broadcast audio library stores clean broadcast audio obtained through track splitting, and the background noise audio library stores background noise audio obtained through track splitting. The volume adjustment module calls the corresponding broadcast audio and background noise audio segments according to different driving scenarios of the simulated driving vehicle, and adjusts the volume of broadcast sound and background noise according to the collected in-vehicle dynamic sound pressure level to simulate the effect of broadcast sound attenuation with distance and environmental changes.

[0010] In the visual simulation subsystem, the full-scene visual simulation module models the target test road section at a 1:1 scale based on the open-source autonomous driving simulation platform (CARLA), including three-dimensional models of roads, traffic markings, traffic signs, guardrails, cameras, and loudspeakers, and sets background traffic flow parameters according to actual traffic conditions; the display output module includes two computer screens, one of which displays the driver's forward field of view in the simulation scene, and the other displays the driver's perspective view in the real video data corresponding to the simulation scene.

[0011] The driving simulator interaction subsystem, namely the driving operation peripheral module, includes a steering wheel, pedals, and gear shifter. It is connected through hardware wiring and driver installation to collect the operation commands of the test driver. The Python program integrated control subsystem realizes the interface configuration and linkage control between the external device and the simulated vehicle through the application programming interface (API) of the open-source autonomous driving simulation platform (CARLA) and the Python multimedia library (Pygame).

[0012] The Python program integrated control subsystem also includes a synchronization and linkage module. This module uses Python programming language code to synchronize data between the acoustic simulation subsystem and the visual simulation subsystem. It binds the loudspeakers for different driving scenarios to the corresponding broadcast audio library of the acoustic simulation subsystem, imports the in-vehicle dynamic sound pressure level collected for the corresponding scenario, and controls the broadcast volume based on the real-time acquired position, speed, and environment of the simulated vehicle by calling the corresponding sound pressure level. It also links the synchronized simulation unit with the driving simulator interaction subsystem, allowing hardware peripherals to control the simulated vehicle's operation and drive real-time acoustic and visual synchronous output. All the running code of the Python program integrated control subsystem is integrated into a single Python test script. For different driving scenarios, by modifying the corresponding parameters in this script and running it, the audio-visual driving scenario simulation test system for highway mainline traffic broadcasting is launched.

[0013] The audio and video data of this system are collected from real vehicle road tests. Target test sections are selected on the main line of the highway, and loudspeakers are installed along the roadside with a spacing of 30-50m, an installation height of 3m, a loudspeaker tilt angle α of 10°, a broadcast transmission power of 120W, a horizontal sway angle θ of 70°-75°, and an overlap rate of ≥30% between adjacent broadcast coverage areas. Data collection vehicles drive through the broadcast coverage area at constant speeds of 30km / h (low speed), 60km / h (medium speed), and 100km / h (high speed), recording audio data inside the vehicle and simultaneously photographing the road conditions ahead and sound level meter data.

[0014] The comprehensive testing process of the system is as follows: the tester sets the corresponding parameters in the Python test script according to the specific driving scenario and then runs the script. After starting the audio and video driving scenario simulation test system of the highway mainline traffic broadcast, the tester enters the driving simulation cockpit for acoustic and visual calibration to confirm that each subsystem is operating normally. The test driver operates the driving simulator, and the system records the operation and feedback data synchronously. After the test, the test driver fills out a subjective evaluation form.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: By constructing an acoustic simulation subsystem, a visual simulation subsystem, and a driving simulator interaction subsystem, this invention forms a modular audio-visual joint simulation testing architecture. Addressing common problems in existing technologies, such as the lack of audio-visual joint simulation, failure to consider complex environments like dynamic traffic flow, and limited testing dimensions, this invention constructs a realistic simulation scenario of a highway mainline segment to accurately and multidimensionally test the audio-visual playback quality and coverage of traffic broadcasts. This system replaces the traditional field testing mode, reduces testing costs, improves testing efficiency, avoids the safety hazards of field testing, and effectively ensures the reliability of traffic broadcast information transmission and driving safety. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the overall system structure of the present invention;

[0017] Figure 2 This is a schematic diagram of the tilt angle of the broadcast speaker of the present invention;

[0018] Figure 3 This is a schematic diagram of the horizontal sway angle of the loudspeaker of this invention;

[0019] Figure 4 This is a schematic diagram of the audio-visual driving scenario simulation test system for traffic broadcasting on the main line of highways according to the present invention. Detailed Implementation

[0020] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] The present invention provides an audio-visual driving scenario simulation test system for traffic broadcasting on the main line of highways, which includes an acoustic simulation subsystem, a visual simulation subsystem, and a driving simulator interaction subsystem.

[0022] I. Acoustic Simulation Subsystem

[0023] The acoustic simulation subsystem includes a broadcast audio library, a background noise audio library, and a volume adjustment module. The broadcast audio library stores clean broadcast audio data obtained after track splitting, while the background noise audio library stores background noise audio data obtained after track splitting. During driving scenario simulation testing, the volume adjustment module calls upon corresponding broadcast audio and background noise audio segments according to different driving scenarios of the simulated vehicle, and adjusts the volume of the broadcast sound and background noise based on the collected in-vehicle dynamic sound pressure level, simulating the effect of broadcast sound attenuation with distance and environmental changes.

[0024] II. Visual Simulation Subsystem

[0025] The visual simulation subsystem comprises a full-scene visual simulation module and a display output module. The full-scene visual simulation module uses the open-source autonomous driving simulation platform (CARLA) to create a 1:1 model of the target test road segment, including 3D models of roads, traffic markings, traffic signs, guardrails, cameras, and loudspeakers. It also sets parameters such as background traffic speed, traffic density, and vehicle arrival frequency based on the actual traffic conditions of the target test road segment. The display output module includes two computer monitors: one displays the driver's forward field of view in the simulation scenario, and the other displays the driver's perspective view from real video data corresponding to the simulation scenario.

[0026] III. Driving Simulator Interaction Subsystem

[0027] The driving simulator interaction subsystem is the driving operation peripheral module. This module includes a steering wheel, pedals, and gear shifter, assembled via hardware wiring and connected with corresponding drivers to collect the driver's operating commands. The Python program-integrated control subsystem uses the application programming interface (API) of the open-source autonomous driving simulation platform (CARLA) and the Python multimedia library (Pygame) to configure and control the interface between external devices and the simulated vehicle.

[0028] IV. Python Program Integrated Control Subsystem

[0029] The Python program integrated control subsystem downloads the application programming interface (API) library of the open-source autonomous driving simulation platform (CARLA) and the Python multimedia library (Pygame) for connecting external devices in the Python programming language environment. It writes code in the Python programming language and manipulates the external devices in the Python multimedia library (Pygame) window to control the driving of vehicles in the open-source autonomous driving simulation platform (CARLA).

[0030] V. System Synchronization and Data Acquisition

[0031] The synchronous linkage module, implemented with Python code, is integrated into the control subsystem via a Python program. This enables data synchronization between the acoustic simulation subsystem and the visual simulation subsystem. The loudspeakers for different driving scenarios are bound to the broadcast audio library of their respective acoustic simulation subsystems. The dynamic sound pressure level inside the vehicle, collected for the corresponding scenario, is imported. Based on the real-time acquired position, speed, and environment of the simulated driving vehicle, the sound pressure level at the corresponding location is used to control the broadcast volume. The synchronized simulation unit is linked with the driving simulator interaction subsystem, and the hardware peripherals control the operation of the simulated vehicle, driving the acoustic and visual synchronous output in real time.

[0032] The system's audio and video data are collected from real-vehicle road tests. Target test sections were selected on the mainline of the highway, and loudspeakers were installed along the roadside with a spacing of 30-50m, an installation height of 3m, a loudspeaker tilt angle α (the angle between the loudspeaker axis and the horizontal plane facing the ground) of 10°, a broadcast transmission power of 120W, and a loudspeaker horizontal yaw angle θ (the angle between the loudspeaker axis and the road shoulder perpendicular to the oncoming traffic direction) of 70°-75°. The overlap rate of adjacent broadcast coverage areas was ≥30%, with no signal blind spots. Data collection vehicles drove through the broadcast coverage area at constant speeds of 30km / h (low speed), 60km / h (medium speed), and 100km / h (high speed), recording audio data inside the vehicle using recording equipment. Simultaneously, a high-speed camera and a dashcam captured road conditions ahead of the vehicle and sound level meter data.

[0033] VI. Comprehensive Testing Process

[0034] Testers set corresponding parameters in a Python test script based on specific driving scenarios and then ran the script. After activating the audio-visual driving scenario simulation test system for the highway mainline traffic broadcast, they entered the driving simulator cockpit for acoustic and visual calibration to confirm that each subsystem was functioning normally. The test driver operated the driving simulator, and the system synchronously recorded operation and feedback data, including steering wheel rotation angle, speed, and frequency; brake and accelerator pedal depth, duration, and frequency; and the number of gear shifts. After the test, the test driver filled out a "Broadcast Intelligibility Form" to obtain feedback on the broadcast audio-visual effects.

[0035] The "Broadcast Intelligibility Scale" uses a 0-5 level rating system. The specific rating levels are as follows: Level 0: Completely inaudible; Level 1: Hears sound, but cannot understand any words; Level 2: Can understand some words, but cannot understand sentences; Level 3: Can understand the general meaning, but with great difficulty; Level 4: Can understand most of the meaning, with occasional difficulty; Level 5: Completely clear, with no difficulty. The "Broadcast Intelligibility Scale" is shown in Table 1.

[0036] Table 1 Broadcast Intelligibility Table

[0037]

[0038] Example: An example of the audio-visual driving scenario simulation test system for highway mainline traffic broadcasting described in this invention is given, providing the implementation process and test results. However, the scope of protection of this invention is not limited to the following example.

[0039] This embodiment selects a 1km straight section from Jiulongxi Toll Station on Qingshang Expressway to Jiaozhou Toll Station on Shenhai Expressway as the core test section. By systematically exploring the impact of different broadcasting strategies on the effect of traffic broadcasting when vehicles are traveling at three speeds on the expressway: low speed 30km / h, medium speed 60km / h, and high speed 100km / h, the practicality of the simulation test system of this invention is verified.

[0040] This embodiment selects vehicle speed, gender of the broadcast voice, broadcast playback speed, and broadcast warning as variables for the survey scheme. Vehicle speed includes low speed (30 km / h), medium speed (60 km / h), and high speed (100 km / h); broadcast voice gender includes male and female; broadcast playback speed includes 3-4 words per second and 4-5 words per second; broadcast warning includes those with and without alarm sounds. The above variables are combined using a full-factor combination method, resulting in 24 different survey schemes. The details of each survey scheme are shown in Table 2.

[0041] Table 2 Survey Plan

[0042]

[0043] Based on the 24 different survey schemes mentioned above, 52 test drivers were selected to conduct simulation tests using driving simulators. Each test driver had to complete 24 survey schemes, for a total of 1248 simulation tests. Simultaneously, the same 52 test drivers conducted 1248 real-world tests in the same scenarios corresponding to the simulations. After each survey scheme test, the test drivers completed the corresponding "Broadcast Intelligibility Form." The results of processing the data from the simulation and real-world tests are shown in Table 3.

[0044] Table 3 Test Results

[0045]

[0046] The test results table shows that Scheme 6, which involves a female voice uttering 3-4 words per second with an alarm tone at a vehicle speed of 30 km / h, has the highest average intelligibility. Based on the average intelligibility score table of simulation and field tests, the average deviation rate between the two was calculated. The average deviation rate refers to the average absolute value of the relative deviation of each field test data point from the simulation test data. The test results show that the overall average deviation rate between the field test and the simulation test is only 3.1%, fully verifying the accuracy and reliability of the simulation test system of this invention. This demonstrates that the "Audio-Video Driving Scene Simulation Test System for Highway Mainline Traffic Broadcasting" proposed in this invention can faithfully reproduce the actual test results and has the ability to replace traditional highway traffic broadcast field test methods. It significantly reduces test costs and safety risks without affecting the reliability of the evaluation conclusions.

Claims

1. A simulation testing system for audio and video driving scenarios in traffic broadcasting on the main line of a highway, characterized in that, The system includes an acoustic simulation subsystem, a visual simulation subsystem, a driving simulator interaction subsystem, and a Python program integrated control subsystem; The acoustic simulation subsystem includes a broadcast audio library, a background noise audio library, and a volume adjustment module, which performs track-by-track processing and dynamic volume adjustment for broadcast sound and background noise; the visual simulation subsystem includes a full-scene simulation visual module and a display output module, wherein the full-scene visual simulation module is used to construct the full-scene simulation visual display content of the target test road section, and the display output module is used to display the driver's forward field of view in the simulation scene and the driver's perspective in the real video data. The driving simulator interaction subsystem, also known as the driving operation peripheral module, uses the force feedback control signal from the Python program integrated control subsystem to achieve linkage control between the driving operation peripheral module and the simulated vehicle. The Python program integrated control subsystem simultaneously collects the driving status data of the simulated vehicle and the driver behavior data acquired by the driving simulator interaction subsystem.

2. The audio-visual driving scenario simulation test system for highway mainline traffic broadcasting according to claim 1, characterized in that, In the acoustic simulation subsystem, the broadcast audio library stores clean broadcast audio obtained through track splitting, and the background noise audio library stores background noise audio obtained through track splitting. The volume adjustment module calls the corresponding broadcast audio and background noise audio segments according to different driving scenarios of the simulated driving vehicle, and adjusts the volume of broadcast sound and background noise according to the collected in-vehicle dynamic sound pressure level to simulate the effect of broadcast sound attenuation with distance and environmental changes.

3. The audio-visual driving scenario simulation test system for highway mainline traffic broadcasting according to claim 1, characterized in that, In the visual simulation subsystem, the full-scene visual simulation module models the target test road section at a 1:1 scale based on the open-source autonomous driving simulation platform. This includes three-dimensional models of roads, traffic markings, traffic signs, guardrails, cameras, and loudspeakers, and sets background traffic flow parameters according to actual traffic conditions. The display output module includes two computer screens, one of which displays the driver's forward field of view in the simulation scenario, and the other displays the driver's perspective from the real video data corresponding to the simulation scenario.

4. The audio-visual driving scenario simulation test system for highway mainline traffic broadcasting according to claim 1, characterized in that, The driving simulator interaction subsystem, namely the driving operation peripheral module, includes a steering wheel, pedals, and gear shifter, which are assembled through hardware wiring and connected by installing drivers to collect the operation commands of the test driver; the Python program integrated control subsystem realizes the interface configuration and linkage control between the driving operation peripheral module and the simulated vehicle through the application programming interface of the open-source autonomous driving simulation platform and the Python multimedia library.

5. The audio-visual driving scenario simulation test system for traffic broadcasting on the main line of a highway according to claim 1, characterized in that, The Python program integrated control subsystem also includes a synchronization and linkage module, which uses Python programming language code to synchronize data between the acoustic simulation subsystem and the visual simulation subsystem. It binds the broadcast speakers of different driving scenarios to the broadcast audio library of the corresponding acoustic simulation subsystem, imports the dynamic sound pressure level inside the vehicle collected in the corresponding scenario, and calls the sound pressure level of the corresponding location to control the broadcast volume according to the real-time acquired position, speed and environment of the simulated driving vehicle. It also links the synchronized simulation unit with the driving simulator interaction subsystem, and controls the operation of the simulated vehicle by the hardware peripherals, driving the acoustic and visual synchronous output in real time. All the running code of the Python program integrated control subsystem is integrated into a Python test script. For different driving scenarios, by modifying the corresponding parameters in the script and running the script, the audio and video driving scenario simulation test system for highway mainline traffic broadcasting is launched.

6. The audio-visual driving scenario simulation test system for highway mainline traffic broadcasting according to claim 1, characterized in that, The audio and video data of this system are collected from real vehicle road tests. Target test sections are selected on the main line of the highway, and loudspeakers are installed along the roadside with a spacing of 30-50m, an installation height of 3m, a loudspeaker tilt angle α of 10°, a broadcast transmission power of 120W, a horizontal sway angle θ of 70°-75°, and an overlap rate of ≥30% between adjacent broadcast coverage areas. Data collection vehicles drive through the broadcast coverage area at constant speeds of 30km / h (low speed), 60km / h (medium speed), and 100km / h (high speed), recording audio data inside the vehicle and simultaneously photographing the road conditions ahead and sound level meter data.

7. The audio-visual driving scenario simulation test system for highway mainline traffic broadcasting according to claim 6, characterized in that, The comprehensive testing process of the system is as follows: the tester sets the corresponding parameters in the Python test script according to the specific driving scenario and then runs the script. After starting the audio and video driving scenario simulation test system of the main highway traffic broadcast, the tester enters the driving simulation cockpit to perform acoustic and visual calibration and confirm that each subsystem is operating normally. The test driver operated the driving simulator, and the system synchronously recorded the operation and feedback data. After the test, the test driver filled out a subjective evaluation form.

Images