A method and system for testing an automatic driving function of a vehicle

Abstract

The present invention discloses a method and system for testing the automatic driving function of an automobile. The method includes: building a test scenario database; respectively equipping an unmanned aerial vehicle (UAV) and a test vehicle with induction radars to ensure that a radar signal communication link is established between the induction radar of the UAV and the induction radar of the test vehicle; the test vehicle and the UAV jointly drive into a test section, the UAV real-time collects radar signals and monitors the distance between the UAV and the test vehicle; the UAV real-time collects the trajectory image information and the driving scenario image information of the test vehicle and performs preprocessing; the flight state of the UAV is adjusted in real time to make the UAV fly in company with the test vehicle synchronously; the preprocessed driving scenario image information of the test vehicle is compared and analyzed with the test scenario database in terms of scene features to obtain a test result and form a test report. The test scenario requirements of the test method of the present invention are low, the test scenario coverage is wide, and the UAV can accurately track the test vehicle, ensuring the accuracy and efficiency of the test.

Description

Technical Field

[0001] The present invention relates to the technical field of automotive road testing, and particularly to a method and system for testing the automatic driving function of an automobile. Background Art

[0002] The testing of the automatic driving function is a mandatory test that must be carried out before the intelligent connected vehicle is launched into the market. Its purpose is to verify the performance, safety and stability of the automatic driving system in the real traffic environment. The main content is to make the automatic driving vehicle perform actual automatic driving on public roads to test the automatic driving functions including coping with complex traffic environments, obeying traffic rules, and handling emergencies. The public road test of the automatic driving function of an automobile can comprehensively evaluate the comprehensive performance of the automatic driving system of the automobile, and is an important link to ensure the safe and effective operation of the automatic driving system in the actual traffic environment, and plays a crucial role in the iteration and commercialization of the automatic driving technology of the automobile.

[0003] At present, the method for testing the automatic driving function of an automobile on the road mainly realizes the test of the automatic driving of the intelligent connected vehicle by equipping devices such as target vehicles, roadside detection devices, and on-vehicle radars, and needs to be carried out in set test scenarios such as closed test fields or designated public roads. However, in reality, the scenarios of automatic driving are complex and diverse, and being able to complete the test on open roads is the most realistic test method for the automatic driving function test of an automobile. Some technical personnel have proposed that the method of tracking by an unmanned aerial vehicle (UAV) can be used to realize the road test of the automatic driving function of an automobile, which can reduce the investment in roadside detection devices and target vehicles, and the detection is more accurate and efficient. At present, most UAVs for tracking test vehicles identify the license plate of the test vehicle or other appearance features on the vehicle through the cameras carried by the UAVs to realize the tracking of the test vehicle by the UAV. Such a tracking method has a low error tolerance rate. The UAV is prone to misidentifying the information of the test vehicle and thus mis-tracking, and requires the test vehicle and the UAV to perform tracking tests in specific test scenarios, resulting in low efficiency of the road test.

[0004] Based on this, it is necessary to use a more efficient and accurate method for testing the automatic driving function of an automobile to efficiently complete the road test of the automatic driving function of the vehicle. Summary of the Invention

[0005] One of the purposes of the present invention is to provide a method and system for testing the automatic driving function of an automobile to solve the problem of high requirements for the test scenarios of the UAV flying with the test vehicle in the prior art, so that the test of the automatic driving function of the vehicle can be carried out in more scenarios and the test of the automatic driving function of the vehicle is made more efficient.

[0006] In order to achieve the above purpose, the technical solutions adopted by the present invention are as follows:

[0007] A method for testing the automatic driving function of a vehicle, comprising:

[0008] S1. Select and record the road scene feature information and input it into the database to build a test scene database; an unmanned aerial vehicle (UAV) and a test vehicle are respectively equipped with induction radars, and it is ensured that a radar signal communication link is established between the induction radar of the UAV and the induction radar of the test vehicle;

[0009] S2. The test vehicle and the UAV jointly drive into the test section for building multiple test scenes. The UAV collects the radar signals in real time and monitors the distance data between the UAV and the test vehicle; the UAV collects the test vehicle trajectory image information and the test vehicle driving scene image information in real time, and after preprocessing, saves them;

[0010] S3. Based on the distance data and the test vehicle trajectory image information, the flight state of the UAV is adjusted in real time according to the preset accompanying flight conditions, so that the UAV and the test vehicle fly in synchronization for a long time;

[0011] S4. Compare and analyze the scene features of the preprocessed test vehicle driving scene image information with the test scene database to determine the test scene and form a test report.

[0012] According to the above technical means, induction radars are installed on both the UAV and the test vehicle in the present invention. The purpose of the UAV tracking the test vehicle is realized through the radar signals between the induction radar of the UAV and the induction radar of the test vehicle. The test of the automatic driving function of the vehicle can be realized without equipping other test equipment, and the coverage of the test scene is wide; the present invention uses the radar signal as the basis for the UAV to track the test vehicle, so that the UAV can still accurately identify, track and accompany the test vehicle in a complex road scene, ensuring the accuracy, continuity and effectiveness of the vehicle test process.

[0013] At the same time, the road scene feature information is input into the database to establish a test scene database. By comparing and analyzing the scene features of the preprocessed test vehicle driving scene image information with the test scene database, the test scene can be quickly determined, making the analysis of the automatic driving function test of the vehicle fast and efficient.

[0014] Further, the preset accompanying flight conditions further include: the height distance between the UAV and the roof of the test vehicle in an air unobstructed section is 1 m to 10 m; when the height distance between the UAV and the test vehicle is greater than 10 m or less than 1 m, the flight state of the UAV is adjusted to meet the preset accompanying flight conditions.

[0015] According to the above technical means, when the UAV flies within the range of 1 m to 10 m from the roof of the test vehicle, it can ensure that the UAV can collect the test vehicle trajectory image information and the test vehicle driving scene image information in real time at a better angle and distance, ensuring that the image information is collected clearly and accurately.

[0016] Further, when the height distance between the drone and the roof of the test vehicle is 1 m to 10 m, the accompanying flight height h of the drone is adjusted according to the speed v of the test vehicle:

[0017] When the speed v of the test vehicle is less than 30 km / h, the accompanying flight height h of the drone is adjusted to: 1 m ≤ h < 3 m;

[0018] When the speed 30 km / h ≤ v < 50 km / h of the test vehicle, the accompanying flight height h of the drone is adjusted to: 3 m ≤ h < 4 m;

[0019] When the speed 50 km / h ≤ v < 80 km / h of the test vehicle, the accompanying flight height h of the drone is adjusted to: 4 m ≤ h < 6 m;

[0020] When the speed 80 km / h ≤ v < 100 km / h of the test vehicle, the accompanying flight height h of the drone is adjusted to: 6 m ≤ h < 8 m;

[0021] When the speed v of the test vehicle is greater than or equal to 100 km / h, the accompanying flight height h of the drone is adjusted to: 8 m ≤ h ≤ 10 m.

[0022] According to the above technical means, the greater the driving speed of the test vehicle, the higher the accompanying flight height h of the drone, so that the range of the drone collecting the trajectory image information of the test vehicle and the driving scene image information of the test vehicle is larger, and the adaptability of the drone to sudden road conditions ahead is better.

[0023] Further, it further includes: the drone collects the flight path image information in real time during flight, analyzes and identifies the obstacles in the flight path according to the flight path image information, and adjusts the flight state of the drone in real time to avoid the obstacles.

[0024] According to the above technical means, when encountering an obstacle on the road, it identifies and avoids the obstacle, so that the drone is not damaged, thereby preventing the interruption of the drone from collecting the trajectory image information of the test vehicle and the driving scene image information of the test vehicle in real time, and ensuring that the test of the vehicle's automatic driving function can be carried out continuously and stably.

[0025] Further, the preset accompanying flight conditions include: the straight-line distance L between the drone and the test vehicle under the aerial obstacle section does not exceed 20 m; when the straight-line distance L between the drone and the test vehicle exceeds 20 m, the flight state of the drone is adjusted in time so that the drone avoids the obstacle while meeting the accompanying flight conditions.

[0026] According to the above technical means, when the straight-line distance L between the drone and the test vehicle does not exceed 20 m in the air obstacle section, it can enable the drone to have enough space for obstacle avoidance operations when tracking the test vehicle, avoiding the drone from colliding with obstacles; after the straight-line distance L between the drone and the test vehicle is greater than 20 m, adjusting the flight height and speed of the drone can enable the radar signals between the drone's induction radar and the test vehicle's induction radar to maintain an effective communication link.

[0027] Further, the scene feature comparison and analysis includes: based on the preprocessed test vehicle driving scene image information, extracting the scene features in the test vehicle driving scene image information and matching them with the scene features in the test scene database to determine the test scene of the test vehicle driving scene image information.

[0028] According to the above technical means, through the extraction and matching of scene features, the actual scene where the test vehicle is located can be quickly identified, which is beneficial to improving the processing efficiency of the test vehicle driving scene image information.

[0029] Further, the preprocessing of the test vehicle trajectory image information and the test vehicle driving scene image information includes: identifying and deleting the video frames when the drone flies from the takeoff position to the accompanying position at the start of the test vehicle's driving, and the video frames when the drone hovers above the test vehicle during the test vehicle's driving.

[0030] According to the above technical means, deleting specific video frames can reduce the amount of information processed in the subsequent image information, improve the efficiency of the subsequent image information processing, and at the same time can clear the storage space of the drone to further store more effective information.

[0031] The present invention also provides an automobile automatic driving function test system, adopting the automobile automatic driving function test method described in any one of the above, and the test system includes:

[0032] A first induction radar and a second induction radar, the first induction radar is carried on the test vehicle and is used for emitting radar signals, and the second induction radar is carried on the drone and is used for receiving the radar signals emitted by the first induction radar;

[0033] An analysis and judgment module, which is used for analyzing the radar signals between the first induction radar and the second induction radar, monitoring the distance between the drone and the test vehicle, and judging whether the flight state of the drone needs to be adjusted;

[0034] A drone control module, which is used for adjusting the flight state of the drone in real time according to the distance data, the test vehicle trajectory image information and the preset accompanying conditions, including the accompanying height h, flight speed and flight path of the drone flight;

[0035] An image acquisition module, which is used for the drone to collect the trajectory image information of the test vehicle and the driving scene image information of the test vehicle in real time;

[0036] A preprocessing module, which is used to preprocess the trajectory image information of the test vehicle and the driving scene image information of the test vehicle, and save the preprocessed trajectory image information of the test vehicle and the driving scene image information of the test vehicle;

[0037] A database analysis module, which is used to perform scene feature comparison and analysis on the preprocessed trajectory image information of the test vehicle and the test scene database.

[0038] According to the above technical means, the information interaction between the first induction radar and the second induction radar enables the drone to monitor the distance from the test vehicle in real time and judge the flight state of the drone through the analysis and judgment module, providing a reliable basis for the drone control module to adjust the flight height, speed and path of the drone. At the same time, the image acquisition module and the preprocessing module can collect the trajectory image information of the test vehicle and the driving scene image information of the test vehicle during the flight of the drone in real time and perform preprocessing, providing a reliable basis for subsequent data analysis.

[0039] Further, the image acquisition module includes a first image acquisition module and a second image acquisition module; the first image acquisition module is used to collect the trajectory image information of the test vehicle and the driving scene image information of the test vehicle; the second image acquisition module is used to collect the drone flight path image information and identify obstacles in the drone flight path.

[0040] Further, the image acquisition module is a camera image sensor.

[0041] The beneficial effects of the present invention are as follows:

[0042] 1. In the test method of the present invention, induction radars are installed on both the drone and the test vehicle. Through the radar signals between the drone induction radar and the test vehicle induction radar, the purpose of the drone tracking the test vehicle is achieved. The test of the automobile automatic driving function can be realized without equipping other test equipment, and the coverage of the test scene is wide; the present invention uses radar signals as the basis for the drone to track the test vehicle, enabling the drone to accurately identify, track and accompany the test vehicle in complex road scenes, ensuring the accuracy, continuity and effectiveness of the vehicle test process.

[0043] 2. Enter the road scene feature information into the database to establish a test scene database, and perform scene feature comparison and analysis on the preprocessed driving scene image information of the test vehicle and the test scene database, which can quickly determine the test scene and make the analysis of the automobile automatic driving function test fast and efficient.

[0044] 3. In the test system of the present invention, the information interaction between the first induction radar and the second induction radar enables the drone to monitor the distance from the test vehicle in real time and determine the flight state of the drone through the analysis and judgment module, providing a reliable basis for the drone control module to adjust the flight height, speed, and path of the drone. At the same time, the image acquisition module and the preprocessing module can collect the trajectory image information of the test vehicle and the driving scene image information of the test vehicle during the flight of the drone in real time and perform preprocessing, providing a reliable basis for subsequent data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only partial embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0046] Figure 1 is the flowchart of the method for testing the automatic driving function of the vehicle of the present invention;

[0047] Figure 2 is the schematic diagram of the composition of the test system for the automatic driving function of the vehicle of the present invention;

[0048] Figure 3 is the first schematic diagram of the drone accompanying the test vehicle of the present invention;

[0049] Figure 4 is the second schematic diagram of the drone accompanying the test vehicle of the present invention;

[0050] Figure 5 is the third schematic diagram of the drone accompanying the test vehicle of the present invention;

[0051] Figure 6 is the schematic diagram of the test scene in the construction area in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] The following will illustrate the embodiments of the present invention with reference to the drawings and preferred embodiments. The drawings are only for illustrative purposes and should not be construed as limiting the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the protection scope of the present invention.

[0053] It should be noted that the illustrations provided in the following embodiments only schematically illustrate the basic concept of the present invention. Therefore, only the components related to the present invention are shown in the drawings, rather than being drawn according to the number, shape, and size of the components in actual implementation. The types, quantities, and proportions of the components in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0054] This embodiment provides a method and system for testing the automatic driving function of a vehicle as shown in Figures 1 to 6 the following figure. As shown in Figure 1 the following figure, the method for testing the automatic driving function of a vehicle includes:

[0055] S1. Select and record the road scene feature information and input it into the database to build a test scene database; the unmanned aerial vehicle (UAV) and the test vehicle are respectively equipped with induction radars, and it is ensured that a radar signal communication link is established between the induction radar of the UAV and the induction radar of the test vehicle;

[0056] S2. The test vehicle and the UAV jointly drive into the test section for building multiple test scenes. The UAV collects radar signals in real time and monitors the distance data between the UAV and the test vehicle; the UAV collects the trajectory image information and the driving scene image information of the test vehicle in real time, and saves them after preprocessing;

[0057] S3. Based on the distance data and the trajectory image information of the test vehicle, the flight state of the UAV is adjusted in real time according to the preset accompanying flight conditions, so that the UAV can accompany the test vehicle synchronously for a long time;

[0058] S4. Compare and analyze the scene features of the preprocessed driving scene image information of the test vehicle with the test scene database to determine the test scene and form a test report.

[0059] In this embodiment, induction radars are installed on both the UAV and the test vehicle. The purpose of the UAV tracking the test vehicle is realized through the radar signals between the induction radar of the UAV and the induction radar of the test vehicle. The test of the automatic driving function of the vehicle can be realized without other test equipment, and the coverage of the test scene is wide; the present invention uses radar signals as the basis for the UAV to track the test vehicle, so that the UAV can still accurately identify, track, and accompany the test vehicle in complex road scenes, ensuring the accuracy, continuity, and effectiveness of the vehicle test process.

[0060] At the same time, the road scene feature information is input into the database to establish a test scene database. By comparing and analyzing the scene features of the preprocessed driving scene image information of the test vehicle with the test scene database, the test scene can be quickly determined, making the analysis of the automatic driving function test of the vehicle fast and efficient.

[0061] Preferably, during road tests, the test vehicle uses various sensors carried on itself, such as cameras, radars, etc., to obtain environmental information. By processing the obtained environmental information, road structure, obstacle information, and vehicle behavior can be extracted, providing a reliable basis for the test of the vehicle's autonomous driving function.

[0062] In this embodiment, the preset accompanying flight conditions include: the height distance between the drone and the roof of the test vehicle is 1m - 10m under the condition of an obstacle-free section in the air; when the height distance between the drone and the test vehicle is greater than 10m or less than 1m, the flight state of the drone is adjusted to meet the preset accompanying flight conditions. The drone flying within the range of 1m to 10m from the roof of the test vehicle can ensure that the drone can collect the test vehicle trajectory image information and the test vehicle driving scene image information in real time at a better angle and distance, ensuring clear and accurate image information collection.

[0063] As Figure 3 or Figure 4 As shown, in this embodiment, when the height distance between the drone and the roof of the test vehicle is 1m - 10m, the accompanying flight height h of the drone is adjusted according to the speed v of the test vehicle:

[0064] When the speed v of the test vehicle < 30 km / h, the accompanying flight height h of the drone is adjusted to: 1m ≤ h < 3m;

[0065] When the speed 30 km / h ≤ v < 50 km / h of the test vehicle, the accompanying flight height h of the drone is adjusted to: 3m ≤ h < 4m;

[0066] When the speed 50 km / h ≤ v < 80 km / h of the test vehicle, the accompanying flight height h of the drone is adjusted to: 4m ≤ h < 6m;

[0067] When the speed 80 km / h ≤ v < 100 km / h of the test vehicle, the accompanying flight height h of the drone is adjusted to: 6m ≤ h < 8m;

[0068] When the speed v of the test vehicle ≥ 100 km / h, the accompanying flight height h of the drone is adjusted to: 8m ≤ h ≤ 10m.

[0069] The greater the driving speed of the test vehicle, the higher the accompanying flight height h of the drone, enabling the drone to collect a larger range of test vehicle trajectory image information and test vehicle driving scene image information, and the drone has better adaptability to sudden road conditions ahead.

[0070] In this embodiment, it further includes: during the flight of the drone, it collects flight path image information in real time, analyzes and identifies obstacles in the flight path based on the flight path image information, and adjusts the flight state of the drone in real time to avoid obstacles. When encountering an obstacle, it identifies and avoids the obstacle, preventing the drone from being damaged, thereby preventing the interruption of the real-time collection of the trajectory image information and the driving scene image information of the test vehicle by the drone, and ensuring that the test of the vehicle's automatic driving function can be carried out continuously and stably.

[0071] As Figure 5 shown, in this embodiment, the preset accompanying flight conditions further include: when the straight-line distance L between the drone and the test vehicle in the air obstacle section does not exceed 20 m; when the straight-line distance L between the drone and the test vehicle exceeds 20 m, the flight state of the drone is adjusted in time so that while the drone avoids obstacles, it meets the accompanying flight conditions. Keeping the straight-line distance L between the drone and the test vehicle not exceeding 20 m in the obstacle section enables the drone to have enough space for obstacle avoidance operations when tracking the test vehicle, avoiding the drone from colliding with obstacles; adjusting the flight height and speed of the drone after the straight-line distance L between the drone and the test vehicle is greater than 20 m can keep the radar signals between the drone's sensing radar and the test vehicle's sensing radar in an effective communication link.

[0072] In this embodiment, the test of the vehicle's automatic driving function includes but is not limited to the following scenarios:

[0073] Before the test vehicle passes through the culvert, the speed is 100 km / h. When passing through the culvert, the speed of the test vehicle decelerates from 100 km / h to 60 km / h to pass through the culvert. At this time, the drone control module controls the drone to decelerate and descend in height, so that the accompanying flight height h of the drone is 4 m to 6 m, and it passes through the culvert synchronously with the test vehicle. After the test vehicle and the drone pass through the culvert, the test vehicle accelerates to 100 km / h, and at the same time, the drone control module controls the drone to accelerate and ascend in height.

[0074] When arriving at the toll station, the test vehicle gradually decelerates and queues up to wait to pass through the toll station. At this time, the speed of the test vehicle is less than 30 km / h. The drone control module controls the accompanying flight height h of the drone to be 1 m to 3 m, and the drone hovers above the test vehicle during the queuing process of the test vehicle. At this time, the trajectory image information and the driving scene image information of the test vehicle collected by the drone are invalid video frames.

[0075] When the test vehicle encounters an obstacle ahead, the test vehicle decelerates, changes lanes or stops to avoid it. At this time, the drone control module controls the drone to decelerate and accompany the flight or hover above the test vehicle. At this time, the height distance between the drone and the roof of the test vehicle should still be 1 m to 10 m.

[0076] When actually conducting vehicle tests, tests in other scenarios also need to be carried out, which cannot be exhausted here, and the above scenarios are only for illustration.

[0077] Preferably, according to standard documents such as "Methods and Requirements for On-Road Tests of Automated Driving Functions of Intelligent Connected Vehicles" (GB / T 41798-2022), when conducting tests on automotive automated driving functions, the selected test scenarios should meet the following requirements:

[0078] (1) Natural environment: The test environment is good, without bad weather such as rainfall, snowfall, and hail, the visibility should be not less than 500m, and when the visibility is lower than 500m, the test vehicle is prohibited from conducting any test work; the temperature range is -10°C to 42°C (not limited to weather, road surface temperature, etc.), and the wind speed is lower than 10m / s.

[0079] (2) Road conditions: The test road surface is generally flat and dry asphalt or concrete pavement. Unless there are special test requirements, the test road is in a closed state throughout the test process; the single-lane width is 3.5m to 3.75m.

[0080] (3) Traffic facilities: Traffic signs, markings, traffic lights, etc. in the test scenario should be set according to the requirements of relevant regulatory documents. In addition to the parameter settings recommended in each test method, the randomness of the parameters of traffic equipment within the standard range can also be increased, such as the traffic light switching time, diversification of the same type of signs, etc.

[0081] (4) The traffic signs and markings in the test scenario are clearly visible and meet the requirements of relevant national standards.

[0082] Preferably, according to standard documents such as "Methods and Requirements for On-Road Tests of Automated Driving Functions of Intelligent Connected Vehicles" (GB / T 41798-2022), when conducting tests on automotive automated driving functions, the test process should also meet the following requirements:

[0083] (1) The execution process and test data of the test vehicle scenario test are both stored and analyzed through the preprocessing module, serving as the basis for the test vehicle's ability test evaluation and review.

[0084] (2) The information recorded during the test process should at least include vehicle status records, vehicle control modes, vehicle positioning information and vehicle postures, vehicle speed, acceleration and other motion states, real-time states of environmental perception and response status signals, 360-degree video monitoring of the vehicle exterior, video monitoring of the vehicle interior, remote control instructions received by the vehicle, manual intervention situations, etc.

[0085] (3) Before the test, the vehicle shall be checked for compliance according to the vehicle test parameter table. Depending on the test route scenario layout, some scenarios can be combined for testing; during the test, different test scenarios need to be combined for testing to evaluate the comprehensive autonomous driving ability of the test vehicle.

[0086] (4) The test vehicle shall conduct all specified scenario tests in one go; during the test, each test scenario shall only be conducted once according to the test method regulations. If the test vehicle fails to meet the requirements of any test scenario, the test shall be terminated.

[0087] Preferably, the test items during the vehicle test include but are not limited to the recognition and response of traffic signs / markings, the recognition and response of the driving status of the vehicle ahead, the recognition and response of pedestrians and non-motor vehicles, etc. Different test items are carried out in multiple test scenarios according to relevant standard requirements.

[0088] In this embodiment, the scenario feature comparison and analysis include: based on the preprocessed test vehicle driving scenario image information, extracting the scenario features in the test vehicle driving scenario image information and matching them with the scenario features in the test scenario database to determine the test scenario of the test vehicle driving scenario image information. Through the extraction and matching of scenario features, the actual scenario where the test vehicle is located can be quickly identified, which is beneficial to improving the processing efficiency of the test vehicle driving scenario image information.

[0089] Preferably, the scenario features include six major scenario element layers:

[0090] (1) Traffic facility layer, including road layout, road surface condition, traffic guiding facilities, etc.;

[0091] (2) Static facility layer, including street lamp poles, isolation belts, trees, buildings, etc.;

[0092] (3) Temporary facility layer, including facilities temporarily changed due to construction or accidents, etc.;

[0093] (4) Dynamic layer, including the geometric and motion features of traffic participants;

[0094] (5) Environment layer, including environmental conditions such as weather and light;

[0095] (6) Interaction layer, including the communication situation between vehicles and between vehicles and other facilities.

[0096] When conducting scenario feature comparison and analysis, use scenario annotation software to accurately annotate scenario elements such as vehicles, pedestrians, obstacles, and traffic guiding facilities in the test vehicle driving scenario image information, and then use the database analysis module to deeply analyze the preprocessed test vehicle driving scenario image information to evaluate the performance of the vehicle's autonomous driving function in different scenarios.

[0097] In this embodiment, the preprocessing of the test vehicle trajectory image information and the test vehicle driving scenario image information includes: identifying and deleting the video frames when the drone flies from the take-off position to the accompanying flight position at the start of the test vehicle's driving, and the video frames when the drone hovers above the test vehicle during the test vehicle's driving. Deleting specific video frames can reduce the amount of image information processed subsequently, improve the efficiency of subsequent image information processing, and at the same time can clear the storage space of the drone and further store more valid information.

[0098] As Figure 2 shown, this embodiment also provides an automotive automatic driving function test system. Using the above-mentioned automotive automatic driving function test method, the automotive automatic driving function test system includes:

[0099] A first induction radar and a second induction radar. The first induction radar is mounted on the test vehicle and is used to emit radar signals. The second induction radar is mounted on the drone and is used to receive the radar signals emitted by the first induction radar;

[0100] An analysis and judgment module, which is used to analyze the radar signals between the first induction radar and the second induction radar, monitor the distance between the drone and the test vehicle, and judge whether the flight state of the drone needs to be adjusted;

[0101] A drone control module, which is used to adjust the flight state of the drone in real time according to the distance data, the test vehicle trajectory image information and the preset accompanying flight conditions, including the accompanying flight height h, flight speed and flight path of the drone;

[0102] An image acquisition module, which is used for the drone to acquire the test vehicle trajectory image information and the test vehicle driving scenario image information in real time;

[0103] A preprocessing module, which is used to preprocess the test vehicle trajectory image information and the test vehicle driving scenario image information, and save the preprocessed test vehicle trajectory image information and the test vehicle driving scenario image information;

[0104] A database analysis module, which is used to compare and analyze the scene features of the preprocessed test vehicle trajectory image information with the test scenario database.

[0105] The information interaction between the first induction radar and the second induction radar enables the drone to monitor the distance from the test vehicle in real time and judge the flight state of the drone through the analysis and judgment module, providing a reliable basis for the drone control module to adjust the accompanying flight height h, flight speed and flight path of the drone. At the same time, the image acquisition module and the preprocessing module can acquire the test vehicle trajectory image information and the test vehicle driving scenario image information during the drone's flight in real time and perform preprocessing, providing a reliable basis for subsequent data analysis.

[0106] Preferably, a vehicle control module is further included in this embodiment. The vehicle control module is mounted on the test vehicle and is used to control the driving speed, acceleration and driving direction of the vehicle. When an unexpected situation occurs during the driving of the test vehicle, the vehicle control module controls the driving state of the vehicle to avoid traffic accidents.

[0107] In this embodiment, the image acquisition module includes a first image acquisition module and a second image acquisition module; the first image acquisition module is used to acquire the test vehicle trajectory image information and the test vehicle driving scene image information, and the second image acquisition module is used to acquire the UAV flight environment information and identify obstacles in the flight environment.

[0108] Specifically, as Figure 4 shown, there is a driver's seat and a co-driver's seat on the test vehicle. When conducting a road test on the vehicle's autonomous driving function, a safety officer A sits in the driver's seat of the test vehicle and a safety officer B sits in the co-driver's seat. Before the road test, safety officer A sets the driving operation of the test vehicle according to the driving route plan. At the same time, before the road test of the test vehicle, safety officers A and B confirm that the communication link between the first induction radar and the second induction radar is normal and effective, and the function of the UAV image acquisition module is normal. After setting the driving operation of the test vehicle and ensuring that all functions are normal, safety officer A starts the vehicle to enter the test state. At the same time, safety officer B starts the UAV flight through the UAV control module and tracks and accompanies the test vehicle for the road test.

[0109] Specific case:

[0110] As Figure 5 shown, a construction area test is conducted. The test road is a long straight road with at least one lane. Warning signs and cones are placed on the lane according to the construction requirements. The test vehicle starts the autonomous driving function and drives at a constant speed of 30 km / h on the test road and enters the construction area where the cones are placed.

[0111] Judgment criterion: If the test vehicle can decelerate in the construction area, slowly avoid and pass the cones, it is regarded as the autonomous driving function test passing in this scenario.

[0112] In the case, the UAV flies at the same speed as the test vehicle, and the UAV laterally offsets according to the offset of the test vehicle driving route so that the UAV can always be located above the top of the test vehicle. When the test vehicle reaches Figure 5 the third cone from left to right in, in addition to acquiring the test vehicle trajectory image information and the test vehicle driving scene image information, the UAV also quickly takes pictures to form test scene picture information and saves it in real time. The test scene picture information is used as supporting evidence to show that the autonomous driving function test meets the judgment criterion. After passing the construction area test, the test vehicle directly drives into the next test scene.

[0113] In the case, when the test vehicle fails to slow down and avoid the conical barrels in the construction area, Safety Officer A manually operates the vehicle control module to stop the test vehicle. At this time, the drone hovers above the test vehicle, takes pictures of the scene when the vehicle stops, and at the same time, the driving recorder in the test vehicle takes pictures of the operation process of Safety Officer A, which is used as evidence to prove that the autonomous driving function test does not meet the judgment criteria. At this time, the test ends.

[0114] After the test vehicle conducts tests on all test scenarios and the test ends, the trajectory image information of the test vehicle and the driving scene image information collected by the drone in real time are saved, and are matched and analyzed with each test scenario in the test scenario database through the database analysis module to form a test report.

[0115] During the driving process of the test vehicle, the first image acquisition module of the drone collects the trajectory image information of the test vehicle and the driving scene image information of the test vehicle in real time. After being transmitted to the preprocessing module, Safety Officer B operates the preprocessing module to preprocess and save the trajectory image information of the test vehicle and the driving scene image information of the test vehicle; the second image acquisition module of the drone collects the flight environment information of the drone in real time and transmits the collected flight environment information to the drone control module in real time. When there are obstacles in the flight path of the drone, Safety Officer B operates the drone control module to control the flight state of the drone, and timely adjusts the accompanying flight height h, flight speed and flight path of the drone to avoid obstacles. After avoiding the obstacles, Safety Officer B also operates the drone control module to adjust the accompanying flight height h, flight speed and flight path of the drone to be consistent with the driving trajectory of the test vehicle; the second image acquisition module of the drone also conducts real-time collection of the trajectory image information of the test vehicle and the driving scene image information of the test vehicle as a supplement to the trajectory image information of the test vehicle and the driving scene image information collected by the first image acquisition module of the drone, so that the image information collected by the drone is more complete. It is worth mentioning that when the test vehicle changes lanes during the driving process, Safety Officer B operates the drone control module to make the drone generate a lateral displacement so that the drone always flies above the test vehicle.

[0116] During the driving of the test vehicle, the first induction radar emits radar signals, and the second induction radar receives the radar signals emitted by the first induction radar. At the same time, the analysis and judgment module of the UAV analyzes the radar signals to judge whether the flight state of the UAV needs to be adjusted. When the flight state of the UAV needs to be adjusted, the safety officer B operates the UAV control module to timely adjust the flight state of the UAV, so that the straight-line distance L between the UAV and the test vehicle is not greater than 20 m, so that the second induction radar can accurately receive the radar signals emitted by the first induction radar, ensuring that the UAV can accurately track and accompany the test vehicle. Specifically, the height distance between the UAV and the roof of the test vehicle is 1 m to 10 m in the air section without obstacles, and the straight-line distance L between the UAV and the test vehicle is not greater than 20 m in the air section with obstacles.

[0117] When the test vehicle encounters a sudden situation ahead during driving and the test vehicle cannot use the automatic driving function to handle it, the safety officer A operates the vehicle control module to make a safety prediction and timely and effective intervention for the driving of the test vehicle, providing safety guidance along the line, driving protection and on-site disposal of emergencies for the test vehicle, avoiding traffic accidents during the driving of the test vehicle, and ensuring the safety of the test vehicle driving.

[0118] In this embodiment, the image acquisition module is a camera image sensor. The first image acquisition module includes one camera image sensor, which is used to acquire the test vehicle trajectory image information and the test vehicle driving scene image information; the second image acquisition module includes eight camera image sensors, and the eight camera image sensors can acquire the environmental information of the UAV flight 360 degrees, enabling the UAV to identify obstacles in the flight environment and avoid obstacles; at the same time, the eight camera image sensors also perform real-time acquisition of the test vehicle trajectory image information and the test vehicle driving scene image information, as a supplement to the test vehicle trajectory image information and the test vehicle driving scene image information acquired by the first image acquisition module of the UAV, so that the image information acquired by the UAV is more complete.

[0119] Obviously, the above embodiments of the present invention are only examples for clearly explaining the present invention, and are not intended to limit the embodiments of the present invention. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the claims of the present invention.

Claims

1. A method for testing the automatic driving function of an automobile, characterized in that: include: S1. Select and record road scene feature information and enter it into the database to build a test scene database; The drone and the test vehicle are equipped with sensing radars respectively, and ensure that the drone sensing radar and the test vehicle sensing radar establish a radar signal communication link; S2. The test vehicle and the drone drive together into a test section with multiple test scenarios, and the drone collects the radar signal in real time to monitor the distance data between the drone and the test vehicle; The drone collects the image information of the test vehicle's trajectory and the test vehicle's driving scene in real time, and saves them after preprocessing; S3, based on the distance data and the test vehicle trajectory image information, adjusting the UAV flight state in real time according to the preset accompanying flight conditions, so that the UAV and the test vehicle can fly synchronously for a long time; S4, comparing and analyzing the scene characteristics of the pre-processed test vehicle driving scene image information with the test scene database, determining the test scene, obtaining the test results and forming a test report; The preset accompanying flight conditions include: the height distance between the UAV and the roof of the test vehicle in the air barrier-free section is 1m to 10m; When the height distance between the UAV and the test vehicle is greater than 10m or less than 1m, adjust the flight state of the UAV to meet the preset accompanying flight conditions; When the height distance between the UAV and the roof of the test vehicle is 1m to 10m, the accompanying flight height h of the UAV is adjusted according to the speed v of the test vehicle: When the test vehicle speed v is less than 30km / h, the UAV's accompanying flight height h is adjusted to: 1m≤h<3m; When the test vehicle speed is 30km / h≤v<50km / h, adjust the UAV's accompanying flight height h to: 3m≤h<4m; When the test vehicle speed is 50km / h≤v<80km / h, adjust the UAV's accompanying flight height h to: 4m≤h<6m; When the test vehicle speed is 80km / h≤v<100km / h, adjust the UAV's accompanying flight height h to: 6m≤h<8m; When the test vehicle speed v≥100km / h, adjust the UAV's accompanying flight height h to: 8m≤h≤10m.

2. The method for testing the automatic driving function of a vehicle according to any one of claim 1, characterized in that: Also includes: During the flight of the UAV, flight path image information is collected in real time, obstacles in the flight path are identified based on the analysis of the flight path image information, and the flight state of the UAV is adjusted in real time to avoid obstacles and accompany the test vehicle.

3. The method for testing the automatic driving function of an automobile according to claim 2, characterized in that: The preset accompanying flight conditions also include: the straight-line distance L between the UAV and the test vehicle does not exceed 20m in the aerial obstacle section; when the straight-line distance L between the UAV and the test vehicle exceeds 20m, the flight state of the UAV is adjusted in time so that the UAV avoids obstacles while meeting the accompanying flight conditions.

4. The method for testing the automatic driving function of an automobile according to claim 1, characterized in that: The scene feature comparison and analysis includes: based on the preprocessed test vehicle driving scene image information, extracting scene features in the test vehicle driving scene image information, and matching them with scene features in a test scene database to determine the test scene of the test vehicle driving scene image information.

5. The method for testing the automatic driving function of a vehicle according to claim 1, characterized in that: Preprocessing the test vehicle trajectory image information and the test vehicle driving scene image information includes: identifying and deleting video frames of the drone flying from a take-off position to a companion flight position when the test vehicle starts to drive, and video frames of the drone hovering above the test vehicle during the test vehicle's driving process.

6. A vehicle automatic driving function test system, characterized in that: The method for testing the automatic driving function of an automobile according to any one of claims 1 to 5 is adopted, wherein the testing system comprises: A first sensing radar and a second sensing radar, wherein the first sensing radar is mounted on the test vehicle and is used to transmit radar signals, and the second sensing radar is mounted on the drone and is used to receive radar signals transmitted by the first sensing radar; An analysis and judgment module, used for analyzing the radar signal between the first sensing radar and the second sensing radar, monitoring the distance between the UAV and the test vehicle, and judging whether the flight state of the UAV needs to be adjusted; A UAV control module is used to adjust the flight state of the UAV in real time according to the distance data, the test vehicle trajectory image information and the preset accompanying flight conditions, including the accompanying flight height h, flight speed and flight path of the UAV flight; An image acquisition module is used for the drone to collect the test vehicle's trajectory image information and the test vehicle's driving scene image information in real time; A preprocessing module, used for preprocessing the test vehicle track image information and the test vehicle driving scene image information, and saving the preprocessed test vehicle track image information and the test vehicle driving scene image information; The database analysis module is used to compare and analyze the scene features of the preprocessed test vehicle trajectory image information with the test scene database.

7. The automobile automatic driving function test system according to claim 6, characterized in that: The image acquisition module includes a first image acquisition module and a second image acquisition module; the first image acquisition module is used to acquire test vehicle trajectory image information and test vehicle driving scene image information; the second image acquisition module is used to acquire drone flight path image information and identify obstacles in the drone flight path.

8. The automobile automatic driving function test system according to claim 7, characterized in that: The image acquisition module is a camera image sensor.

Images