Vehicle machine application test scene building and testing method for on-duty and off-duty specific road sections

Abstract

The invention relates to the technical field of intelligent network connection vehicle-to-vehicle infotainment application testing, and discloses an on-off specific road section vehicle-to-vehicle infotainment application testing scene building and testing method comprising the following steps: converting a CAN bus signal and an ADB instruction of a vehicle-to-vehicle infotainment into a cloud recognizable communication protocol through a 4G / 5G network conversion module of a cloud testing platform; deploying a signal test terminal on a test vehicle, connecting a USB interface and a CAN interface of the vehicle machine, and collecting state data of the vehicle machine in real time; based on historical traffic data, test routes of typical on-off scenes including peak congestion road sections, tunnels and viaducts are planned, and longitude and latitude coordinate sequences of the routes are obtained. According to the invention, automatic testing in a drive test environment can be realized through the cloud testing platform, the technical blank of laboratories and drive test scenes is filled, and the testing efficiency and accuracy are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent networked vehicle machine application testing, and in particular to a method for building and testing a specific route vehicle machine application test scene for commuting. BACKGROUND

[0002] The machine application testing of intelligent networked vehicles involves complex scenarios in multiple dimensions such as network environment, geographic location, and traffic conditions. Existing testing solutions are mainly divided into two categories: laboratory automation testing and road testing manual testing: Laboratory testing: Through hardware-in-the-loop (HIL) devices and simulation software to simulate signal interaction, it can achieve automatic script execution, but it cannot truly reproduce network fluctuations, geographic location dependence (such as latitude and longitude), and dynamic traffic environment in real-time road conditions.

[0003] Manual testing on the road: It relies on test personnel to manually operate in actual road environments (such as manual listening for responses in vehicle-mounted voice testing), making it difficult to cover extreme scenarios (such as low battery interaction on mobile phones) and lacking in data recording accuracy (such as human error).

[0004] For the above two types of testing, the current technical solutions have the following defects: 1. Low testing efficiency: Laboratories cannot simulate real-time road conditions, and during road testing, they cannot connect to laboratory servers, requiring manual one-by-one clicking operations (such as the bench system disclosed in patent CN113049272A requiring manual configuration of wiring harnesses), especially during peak commuting hours, repeated testing is time-consuming and labor-intensive.

[0005] 2. Low testing coverage: Laboratories have difficulty simulating complex network environments (such as 4G / 5G signal attenuation), dynamic latitude and longitude changes, and server bandwidth pressure (such as the base station simulation disclosed in patent CN114756009A only supporting fixed network simulation), making it difficult to cover extreme scenarios (such as multi-vehicle interaction and tunnel signal interruption) and increasing the risk of missed detection.

[0006] 3. Insufficient testing accuracy: Manual recording cannot capture multiple dimensions of data such as instruction request time, response delay, and latitude and longitude coordinates (such as the response time testing disclosed in patent CN115560994A relying on manual comparison), and the data accuracy is low (such as ±10ms error), making it difficult to reproduce the problem.

[0007] 4. High labor costs: Professional operation hardware (such as ADB touch simulation disclosed in patent CN112416775A needs manual debugging), each version iteration needs to repeat manual testing, and the utilization rate of hardware resources is low (such as the test bench disclosed in patent CN113049272A only supports a single vehicle model).

[0008] The prior art cannot fuse the automatic test script and the multi-dimensional environment simulation under the real-time road condition, resulting in frequent compatibility and stability problems of the car machine application in the real scene. Therefore, it is necessary to study a new car machine application test method. SUMMARY

[0009] The present application aims to provide a specific route car machine application test scene building method for commuting, which provides scene support for the new car machine application test method, to solve the problem that the prior art cannot fuse the automatic test script and the multi-dimensional environment simulation under the real-time road condition, resulting in frequent compatibility and stability problems of the car machine application in the real scene.

[0010] To solve the above problems, the present application adopts the following technical solutions: Scheme one: a specific route car machine application test scene building method for commuting, comprising: Step one, through the 4G / 5G network conversion module of the cloud test platform, convert the CAN bus signal and ADB instruction of the car machine into a communication protocol recognizable by the cloud; Step two, deploy a signal test terminal on the test vehicle, connect the car machine USB interface and CAN interface, and collect car machine state data in real time; Step three, based on historical traffic data, plan a test route containing a typical commuting scene of peak congestion section, tunnel and viaduct, and obtain the latitude and longitude coordinate sequence of the route.

[0011] Advantages: The cross-network communication between the laboratory server and the road test vehicle is realized, breaking through the laboratory network limitation of the prior art and supporting real-time data backhaul; the standardized test route covers more than 90% of typical road conditions, reducing the time consumption of manual route planning by more than 30%.

[0012] The present application realizes automatic testing in the road test environment through the cloud test platform, fills the technical gap between the laboratory and the road test scene, improves the test efficiency and accuracy, and reduces the labor cost.

[0013] Preferably, the 4G / 5G network conversion module supports dynamic signal attenuation simulation, which reproduces the weak network environment in the building shelter and tunnel by adjusting the signal strength parameter in the range of-100dBm to-50dBm.

[0014] Advantages: Compared with the prior art pressure test, the signal strength cannot be dynamically adjusted, the application can simulate a real weak network scene, and the test coverage is increased to more than 95%.

[0015] Preferably, the signal test terminal integrates a high-precision GPS module, and the positioning accuracy is less than or equal to 1 m; the latitude, longitude and timestamp of the test vehicle are synchronized in real time to the cloud test platform.

[0016] Beneficial effects: Compared with the traditional GPS accuracy, sub-meter positioning data is provided to ensure the accuracy of the geographic location dependent test.

[0017] Preferably, in step two, environmental sensors are also deployed inside the test vehicle to collect in-vehicle noise decibel values, mobile phone power states, simulate 60dB-90dB high noise, and simulate low power extreme interaction scenarios with a proportion of less than 20%.

[0018] The "low battery interaction of the mobile phone" and other artificial difficult-to-simulate scenarios are reproduced, and the test case coverage is increased by 20%.

[0019] Scheme two: the application also provides a test method for a specific road section vehicle-machine application test scene during commuting, which adopts the scene building method of scheme one to complete test scene building, including: Step 1: the cloud test platform issues an automatic test script to the Agent terminal of the test vehicle, and the script contains a simulated user operation navigation route planning and a voice instruction wake-up; Step 2: the Agent terminal executes the script, controls the vehicle-machine interface interaction through ADB instructions, and collects vehicle-machine response data including instruction delay and interface jump time through CAN bus; Step 3: real-time feedback of test data including vehicle-machine response data to the cloud test platform to generate a multi-dimensional test report containing timestamp, latitude, longitude and signal strength.

[0020] Beneficial effects: Compared with manual operation, the efficiency of automatic execution is increased by 80% compared with manual test, and the data record dimension is increased by more than 50%.

[0021] Preferably, the automatic test script in step 1 supports multi-task parallel execution, and simultaneously tests the compatibility of navigation path planning and music playing functions, monitors CPU occupancy rate ≤ 85%, and memory leakage ≤ 50MB.

[0022] Beneficial effects: Compared with the prior art multi-module independent test, concurrent test covers complex scenarios, and the hardware resource utilization rate is increased by 40%.

[0023] Preferably, in a weak network scenario, the cloud measurement platform dynamically adjusts the bandwidth fluctuation of the simulation server at 1Mbps to 10Mbps, tests the breakpoint continuation and cache loading functions of the car machine application.

[0024] Advantages: Accurately reproduce the "high concurrency bandwidth pressure" scenario not covered by existing technologies, and the defect discovery rate is increased by 30%.

[0025] Preferably, the AI algorithm is used to analyze the test report, automatically identify abnormal points including instructions with response time> 500ms, and associate environmental data including latitude, longitude and signal strength to generate problem reproduction paths.

[0026] Advantages: Compared with the manual analysis efficiency in the prior art, the problem positioning time is shortened from 2 hours to 15 minutes.

[0027] Preferably, the cloud measurement platform remotely restarts the car machine and switches the 4G / 5G network mode to support remote control of the hardware state of the test vehicle.

[0028] Advantages: Compared with the time-consuming wire replugging in the prior art, remote maintenance reduces hardware operation cost by more than 50%.

[0029] Preferably, the multi-dimensional test report automatically generated after the test is a visual report, including a heat map showing the test pass rate of each section and a waterfall chart showing the instruction response time distribution.

[0030] Advantages: Compared with the manual report in the prior art, the report generation efficiency is improved by 90%, and the data visualization degree is improved by 60% compared with the traditional text report.

[0031] The advantages of the present scheme are: 1. Efficiency improvement: 80% of manual operations are replaced by automatic script execution, and the time-consuming of single-scene testing is shortened from 2 hours to 20 minutes; Support 7x24 unattended testing (such as night peak period simulation), and the device utilization rate is improved by 3 times.

[0032] 2. Comprehensive coverage: Fusion of real-time road conditions, network environment, and hardware state multi-dimensional simulation, test case coverage from 500 to 1500; Support for automatic testing of extreme scenarios (such as -100dBm signal, 0% phone power), and the coverage rate of scenarios that are difficult to reproduce manually is 100%.

[0033] 3. Data accuracy: Sub-millimeter positioning, millisecond-level timestamp and signal strength real-time synchronization, data error ≤5%; Automatic report generation of multi-dimensional analysis charts, problem positioning efficiency improved by 70%.

[0034] 4. Cost reduction: Reduce more than 50% of professional testers, reduce training costs by 60%; Hardware resource sharing (such as cloud multi-vehicle parallel testing), equipment procurement cost reduced by 40%. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 The structure diagram of the first embodiment of the present application.

[0036] Figure 2 The structure diagram of the second embodiment of the present application.

[0037] Figure 3 The connection structure diagram of the Agent terminal in the vehicle machine test process. DETAILED DESCRIPTION

[0038] The following will be further described in detail through specific embodiments: The application test scene building and test method of the vehicle machine application in the specific road section of commuting, With the continuous development of intelligent networked vehicles, vehicle machine applications are constantly enriched. Due to the dependence of applications on network environment, dependence of applications on geographic location (latitude and longitude coordinates), dependence of applications on vehicle congestion in test road section (commuting period), dependence of applications on vehicle machine application server bandwidth, etc., the dependence of test cases is becoming more and more complex. It is very difficult to test in a laboratory simulation test environment. Therefore, scene testing under real-time road conditions is needed.

[0039] The test system of the present application is a cloud test platform independently developed by our company, including a test server, a use case execution Agent and a 4G / 5G network conversion module. The platform realizes connection with the test server of the laboratory, execution of automatic scripts and realization of automatic testing in complex environment during road testing.

[0040] System principle: each module of the above cloud test platform realizes the following functions respectively: Test server: used for recording, editing and executing automated test scripts, reducing the workload of manual testing; support simulating user operations, covering various usage scenarios such as login, payment, navigation, etc.; support timing test tasks, convenient for night or non-working hours testing; provide a large number of real devices (different brands, models, operating system versions) for testing to ensure the compatibility of the App on different devices; test the display effect of the App under different screen resolutions and sizes; test the startup time of the App to optimize user experience; monitor the resource consumption of the App such as CPU, memory, power, network traffic, etc.; simulate high-concurrency user scenarios to test the stability of the App under high load, etc.

[0041] Agent terminal: refers to a terminal device installed with intelligent agent (Agent) software, which is used in Figure 1 and Figure 2 to refer to a converter for registering test devices to the cloud platform to ensure that the device can be identified and managed by the platform; real-time monitoring of the running state of the device, including power, network connection, CPU / memory usage, etc.; support remote control of the device, such as reboot, install / uninstall App, clean cache, etc.; receive test tasks from the cloud test platform, including automated test scripts, manual test instructions, etc.; execute automated test scripts on the device, simulate user operations (such as clicking, sliding, inputting, etc.); according to the platform instructions, reasonably arrange the execution order and time of the test tasks; monitor and collect App performance data such as startup time, CPU / memory occupation, network traffic, etc.; parse the automated test scripts issued by the cloud test platform; run test tasks on devices of different models, resolutions, and operating systems to ensure the compatibility of the App; ensure that the Agent module has the necessary permissions on the device; detect abnormal conditions during testing, such as App crashes, device disconnections, etc.

[0042] 4G / 5G communication network module: simultaneously supports 4G / 5G signal to ADB instruction and 4G / 5G signal to CAN bus instruction.

[0043] The present application realizes automatic testing in the environment of automobile road test, and realizes the execution of automatic cases that can only be tested in the laboratory in real-time road conditions, which is more efficient. The automatic testing in various complex scenarios is more accurate, saves manpower cost, improves test quality, improves device utilization, and improves more accurate test reports.

[0044] As shown in Figure 3 , during the test process, the Agent terminal is used as a test terminal for ADB / CAN signal to 4G / 5G in the open environment of the application test process on the specific road section to work, solving the problem of limited test environment and insufficient real-time in the prior art, realizing remote and efficient CAN bus signal testing and monitoring.

[0045] An agent terminal, comprising: An ADB signal receiving module: for receiving ADB signals from the device under test; supports multiple connection methods such as USB, Wi-Fi, etc.

[0046] A CAN bus receiving module: for receiving signals from the CAN bus; supports standard CAN (ISO 11898) and CANFD (Flexible Data Rate) protocols, and provides high anti-interference capability, suitable for complex electromagnetic environments.

[0047] A 4G / 5G communication module: converts the converted signals and transmits them to a remote server or test platform through a 4G / 5G network; supports multiple frequency bands and multiple standards (such as LTE, NR), and adapts to different network environments.

[0048] A control and configuration module: provides a user interface or API interface for configuring test parameters (such as baud rate, frame format, etc.); supports remote control and real-time status monitoring.

[0049] A power management module: provides efficient power management functions, supports long-term operation; supports battery power supply and external power supply adaptation.

[0050] In operation, the test vehicle is connected to the test terminal through the ADB interface. The agent terminal receives the ADB signal and converts it to a data format suitable for 4G / 5G network transmission through the signal conversion module built-in the control and configuration module, and adds a field identifying this message as ADB type.

[0051] The CAN bus signal of the test vehicle is connected to the agent terminal through the CAN bus interface module.

[0052] The signal processing and conversion module in the control and configuration module analyzes and encrypts the CAN bus signal, converts it to a data format suitable for 4G / 5G network transmission, and adds a field identifying this message as CAN type.

[0053] The data converted by the control and configuration module is sent to the cloud measurement App server through the network base station through the 4G / 5G communication module. The server analyzes and processes the data and feeds back the results to the user.

[0054] The agent terminal not only converts and collects various signals in the open environment vehicle application test, but also provides the connection of various test devices and test platforms for testing without affecting the various signal interferences existing in the open environment itself, making it possible to perform vehicle application testing in an open environment.

[0055] Example one AsFigure 1 As shown, this embodiment is a navigation application test under heavy traffic and road conditions.

[0056] I. Scene building: Plan the evening peak test route (such as the main road of the city ring line, including 3 congestion points and 2 viaducts), and determine the congestion period as 17:30-19:00 through historical traffic data; Deploy signal test terminal (integrated GPS module, positioning accuracy 0.8m), connect car machine USB and CAN interface, 4G / 5G network conversion module set signal strength to-80dBm (simulate building obstruction); The in-vehicle environmental sensor collects noise of 75dB (simulates real traffic noise).

[0057] The test terminal connects the car machine through USB / Wi-Fi, accesses the bus through CAN interface, and collects ADB / CAN signals in real time; the 4G / 5G module converts the signal into cloud protocol, adds latitude and longitude, timestamp, etc. Label and transmit to cloud test platform; the platform generates test route according to historical traffic data, dynamically adjusts the signal attenuation parameters of the terminal, and simulates the network environment of real road conditions.

[0058] Through the building method of this embodiment, the route building time is shortened from 2 hours manually to 30 minutes (automated planning + parameter remote configuration); the environmental simulation dimension is increased from 3 (signal strength, time, and location) to 6 (new electromagnetic interference, vehicle density, and server bandwidth), and the scene reality is improved by 60%.

[0059] II. Automated test execution: The cloud test platform issues a navigation test script: simulate user input "home" address, select "avoid congestion" route, and switch music playback in the middle; The agent terminal executes the script, records the navigation path calculation time (average 2.3s, ≤3s is qualified), and the map loading delay (average 450ms, ≤500ms is qualified); Real-time data is returned to the cloud to generate a heat map containing latitude and longitude, and the navigation response delay of the congestion section is increased to 800ms (triggering a yellow warning).

[0060] The cloud measurement platform issues an automatic script to a test terminal, the terminal executes car machine operation (such as voice wake-up, navigation search), and simultaneously collects response data (such as audio recognition accuracy, map loading time consumption); the data is real-time returned after encryption, the platform analyzes abnormal points through an AI algorithm, and generates a reproduction path by associating environmental data; the test terminal supports remote restart of the car machine or switching of a network mode, and automatically recovers a test process. Through the test method of the application, the automatic test coverage rate is improved from 50% to 90%, the number of manual interventions is reduced by 85%, the problem reproduction rate is improved from 40% to 90%, and the average positioning time is shortened from 4 hours to 40 minutes.

[0061] Result analysis: The map loading failure problem in signal switching under the viaduct is found, the signal strength-95dBm data is associated, the defect of the cache mechanism in the weak network environment is located, and the delay is reduced to 480ms after optimization.

[0062] Example two As shown in the figure, this embodiment is a voice interaction test in a weak network environment. Figure 2

[0063] I. Scene building: The 4G / 5G network conversion module simulates signal strength-100dBm (close to no signal) in the tunnel entrance section; The phone battery is set to 15% (low power mode), and the in-vehicle noise is 85dB (simulating a construction section).

[0064] II. Automatic test execution: The cloud measurement platform issues a voice test script: wake up the car machine for 3 times in succession, and the instructions are "play a song", "navigate to the company" and "lower the volume"; The agent terminal outputs voice instructions (volume 75dB) through the artificial mouth, the sound collector collects the car machine response audio, and the recognition accuracy ≥95% is qualified; The car machine is monitored whether to use the local voice cache under the weak network, and the response time ≤4s is qualified (the actual test is 3.8s).

[0065] Result analysis: It is found that the car machine does not timely switch to the local cache under the weak network, resulting in a timeout (5.2s) of the first instruction, and after optimization, the local cache triggering delay ≤1s, the response time is stably within 3.5s.

[0066] Compared with the prior art, the application has the following advantages: 1. Technical combination innovation: ​In the existing technology, laboratory automated testing (such as patent CN115550760A) and manual drive testing (such as patent CN113590399A) are independent of each other. However, this invention uses a 4G / 5G network conversion module + cloud testing platform to achieve automated script execution and real-time data feedback in a drive testing environment for the first time, breaking through scenario limitations.

[0067] 2. Dynamic environment simulation: In the existing technology, such as the base station simulation of patent CN114756009A, only fixed network parameters are supported. The present invention reproduces the real network environment that changes with time and space by dynamically adjusting parameters such as signal strength and bandwidth. It is not a simple parameter adjustment, but a multi-dimensional dynamic modeling.

[0068] 3. Data fusion analysis: In the existing technology, such as patent CN112416775A, only a single operation data is recorded. The present invention integrates GPS, environmental sensors and vehicle bus data to achieve "time-space-command-response" four-dimensional correlation analysis, which significantly improves the technical complexity.

[0069] 4. Automated closed loop: Existing technologies require manual intervention in hardware (such as the harness adjustment in patent CN113049272A). The present invention uses an agent terminal to achieve a closed loop of remote hardware control, automatic script execution, and report generation. This is not a simple process automation, but a full-link intelligentization.

[0070] The above is only an embodiment of the present invention, and the common knowledge such as the specific technical solutions and / or characteristics in the solution are not described in detail here. It should be pointed out that for those skilled in the art, without departing from the technical solution of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the description can be used to interpret the content of the claims.

Claims

1. A method for building a vehicle-mounted machine application test scenario for a specific commuting section, characterized in that: include: Step 1: Use the 4G / 5G network conversion module of the cloud testing platform to convert the CAN bus signals and ADB commands of the vehicle computer into a communication protocol that can be recognized by the cloud; Step 2: Deploy a signal test terminal on the test vehicle, connect the vehicle's USB interface and CAN interface, and collect vehicle status data in real time; Step 3: Based on historical traffic data, plan a test route for typical commuting scenarios that includes peak-hour congested sections, tunnels, and overpasses, and obtain the longitude and latitude coordinate sequence of the route.

2. The method for building a vehicle-mounted computer application test scenario for a specific commuting section according to claim 1 is characterized in that: The 4G / 5G network conversion module supports dynamic signal attenuation simulation, and by adjusting the signal strength parameters within the range of -100dBm to -50dBm, it can reproduce the weak network environment of building obstruction and tunnel.

3. The method for building a vehicle-mounted computer application test scenario for a specific commuting section according to claim 1 is characterized in that: The signal test terminal integrates a high-precision GPS module with a positioning accuracy of ≤1m, and synchronizes the latitude and longitude and timestamp of the test vehicle to the cloud testing platform in real time.

4. The method for building a vehicle-mounted computer application test scenario for a specific commuting section according to claim 1 is characterized in that: In step two, environmental sensors are also deployed in the test vehicle to collect the decibel value of the noise inside the vehicle and the battery status of the mobile phone, simulating loud noises of 60dB-90dB and extreme interaction scenarios with low battery power of less than 20%.

5. A method for testing vehicle-mounted computer applications in a specific commuting section, characterized in that: The test scenario is constructed using the scenario construction method according to claim 1, comprising: Step 1: The cloud testing platform sends an automated test script to the test vehicle's agent terminal. The script includes navigation route planning and voice command wake-up that simulates user operations. Step 2: The Agent terminal executes the script, controls the vehicle interface interaction through ADB commands, and collects vehicle response data including command delay and interface jump time through the CAN bus; Step 3: Transmit the test data including the vehicle computer response data to the cloud testing platform in real time, and generate a multi-dimensional test report including timestamp, longitude and latitude, and signal strength.

6. The vehicle-mounted computer application test scenario test method for a specific commuting section according to claim 5 is characterized in that: The automated test script in step 1 supports multi-task parallel execution, and simultaneously tests the compatibility of navigation path planning and music playback functions, monitoring CPU usage ≤ 85% and memory leaks ≤ 50MB.

7. The vehicle-mounted computer application test scenario test method for a specific commuting section according to claim 5 is characterized in that: In a weak network scenario, the cloud testing platform dynamically adjusts the simulated server bandwidth fluctuations from 1Mbps to 10Mbps to test the breakpoint resumption and cache loading functions of the vehicle-mounted application.

8. The vehicle-mounted computer application test scenario test method for a specific commuting section according to claim 5 is characterized in that: Use AI algorithms to analyze test reports, automatically identify anomalies, including instructions with response times greater than 500ms, and correlate environmental data such as latitude and longitude and signal strength to generate a path to reproduce the problem.

9. The vehicle-mounted computer application test scenario test method for a specific commuting section according to claim 5 is characterized in that: Remotely restart the vehicle computer and switch the 4G / 5G network mode through the cloud testing platform, and support remote control of the test vehicle's hardware status.

10. The vehicle-mounted computer application test scenario test method for a specific commuting section according to claim 5 is characterized in that: The multi-dimensional test report automatically generated after the test is completed is a visual report, which includes a heat map showing the test pass rate of each road section and a waterfall chart showing the distribution of command response time.

Images