Unmanned vehicle testing method, device, unmanned vehicle and medium

Abstract

The embodiment of the present invention discloses a test method, device, unmanned vehicle and medium for an unmanned vehicle. The method includes: obtaining test collection data of a vehicle perception system under different test environments; determining correction data based on the test collection data; and establishing a corresponding relationship between the correction data and the test environment. Through the above test method, when the vehicle is driving in different complex environments, the vehicle perception system can directly call the correction data corresponding to the environment based on the collected environmental information data, thereby obtaining environmental information data with less interference, so that the on-board controller can make correct control instructions, and avoid driving accidents when the vehicle is driving in complex environments.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of automobile control technology, and in particular to a test method and device for an unmanned vehicle, an unmanned vehicle, and a medium. Background Art

[0002] At present, with the development of the national economy, the demand for express delivery and instant delivery orders is increasing. The annual express delivery volume is trillions of pieces. The improvement of terminal delivery efficiency and the reduction of delivery costs have become a major problem that the industry urgently needs to solve. In order to solve this pain point in the industry, driverless vehicles have been born.

[0003] Before driverless vehicles are put into use, various driving performance characteristics of the vehicles will be tested. However, since driverless vehicles are emerging products, relevant management departments have not yet issued mandatory standards for the vehicle size, product configuration, and vehicle performance of driverless vehicles, nor are there targeted testing method documents. The sizes of driverless vehicles on the market vary, and the test methods, test equipment, test processes, and evaluation indicators are also different.

[0004] In the existing technology, when major manufacturers conduct actual vehicle tests, they often only test whether the vehicle can complete conventional driving operations such as driving, steering and braking according to the sensors installed on the vehicle body under normal driving conditions. However, due to the complexity of the actual driving environment, the above test methods cannot fully guarantee the safe driving of the vehicle, resulting in many problems in the actual application of driverless vehicles, which makes it impossible for driverless vehicles to be commercialized on a large scale. Summary of the invention

[0005] In view of the problems existing in the above-mentioned prior art, the embodiments of the present invention provide a test method, device, unmanned vehicle and medium for an unmanned vehicle, so as to conduct a comprehensive real vehicle test on the unmanned vehicle and improve the driving stability of the unmanned vehicle in complex environments.

[0006] In a first aspect, an embodiment of the present invention provides a method for testing an unmanned vehicle, the method comprising:

[0007] Obtain test data of vehicle perception systems under different test environments;

[0008] Determine correction data based on test collection data;

[0009] Establish the correspondence between the correction data and the test environment.

[0010] Optionally, the correction data is determined based on the test collection data, including:

[0011] Obtaining baseline acquisition data of the vehicle perception system in a baseline test environment; the baseline test environment is a test environment without additional test conditions;

[0012] Correction data is determined based on the test acquisition data and the reference acquisition data.

[0013] Optionally, after establishing the correspondence between the correction data and the test environment, the following steps are also included:

[0014] Correct the test acquisition data according to the correction data.

[0015] Optionally, before obtaining the test collection data of the vehicle perception system under different test environments, the following is also included:

[0016] Adjust the test environment.

[0017] Optionally, the test environment includes a rain test environment;

[0018] Adjusting the test environment includes:

[0019] Adjust at least one of the pipeline pressure value, the rain time and the pipeline flow rate of the rain test equipment;

[0020] Adjust at least one of the pitch angle, roll angle and heading angle of the vehicle.

[0021] Optionally, the vehicle perception system includes an image acquisition module, which includes an image acquisition unit and a wiper;

[0022] Before obtaining the test data of the vehicle perception system under different test environments, it also includes:

[0023] Adjust the working state of the wiper according to the test environment, and the working state includes the swing amplitude and the swing frequency.

[0024] Optionally, the vehicle perception system includes an image acquisition module, which includes a plurality of image acquisition units arranged around the body of the unmanned vehicle;

[0025] Obtain test data of vehicle perception systems under different test environments, including:

[0026] Acquire test acquisition data of multiple image acquisition units under different test environments;

[0027] Determine the correction data based on the test collection data, including:

[0028] Determine a plurality of correction data according to a plurality of test collection data;

[0029] Determining average correction data based on the plurality of correction data;

[0030] Establish the corresponding relationship between the correction data and the test environment, including:

[0031] Establish the correspondence between the average correction data and the test environment.

[0032] In a second aspect, an embodiment of the present invention further provides a test device for an unmanned vehicle, the test device comprising:

[0033] A test data acquisition module is used to acquire test data of the vehicle perception system under different test environments;

[0034] A correction data determination module, used to determine the correction data according to the test collection data;

[0035] Correspondence establishment module: used to establish the correspondence between the correction data and the test environment.

[0036] In a third aspect, an embodiment of the present invention further provides an unmanned vehicle, the vehicle comprising:

[0037] one or more processors;

[0038] A storage device for storing a plurality of programs,

[0039] When at least one of the multiple programs is executed by one or more processors, the one or more processors implement the unmanned vehicle testing method provided by any embodiment of the present invention.

[0040] In a fourth aspect, an embodiment of the present invention further provides a medium on which a computer program is stored, and when the program is executed by a processor, the method for testing an unmanned vehicle described in any embodiment of the present invention is implemented.

[0041] The technical solution provided by the embodiment of the present invention first obtains the test data collected by each vehicle perception system under different test environments, then corrects the test data, determines the correction data, and finally establishes the relationship between the correction data and the test environment. Through the above method, when the vehicle is driving in different complex environments, the vehicle perception system can directly call the correction data corresponding to this environment based on the collected environmental information data, thereby obtaining environmental information data with less interference, so that the on-board controller can make correct control instructions, avoiding driving accidents when the vehicle is driving in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flow chart of a method for testing an unmanned vehicle provided by an embodiment of the present invention;

[0043] Figure 2 is a flow chart of another unmanned vehicle testing method provided by an embodiment of the present invention;

[0044] Figure 3 is a flow chart of another unmanned vehicle testing method provided by an embodiment of the present invention;

[0045] Figure 4 is a flow chart of another unmanned vehicle testing method provided by an embodiment of the present invention;

[0046] Figure 5 is a flow chart of another unmanned vehicle testing method provided by an embodiment of the present invention;

[0047] Figure 6 It is a structural schematic diagram of a test device for an unmanned vehicle provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present invention clearer, the specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described here are only used to explain the present invention, rather than to limit the present invention.

[0049] It should also be noted that, for ease of description, only the part relevant to the present invention but not all content is shown in the accompanying drawings. It should be mentioned before discussing exemplary embodiments in more detail that some exemplary embodiments are described as processing or methods depicted as flow charts. Although the flow chart describes each operation (or step) as sequential processing, many operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of each operation can be rearranged. When its operation is completed, the processing can be terminated, but it can also have additional steps not included in the accompanying drawings. The processing can correspond to methods, functions, procedures, subroutines, subprograms, etc.

[0050] Figure 1 is a flow chart of a method for testing an unmanned vehicle provided by an embodiment of the present invention. This embodiment can be applied to actual vehicle testing of unmanned vehicles. The method can be executed by a test device for an unmanned vehicle provided by any embodiment of the present invention, which can be implemented in the form of software and / or hardware. Figure 1 , the method specifically comprises the following steps:

[0051] S110, obtaining test collection data of the vehicle perception system under different test environments.

[0052] Since driverless vehicles have no driver to control them, all control commands are issued by the on-board controller, and the basis for the on-board controller to issue control commands is the various information collected by the vehicle perception system. The vehicle perception system refers to the various sensors installed on the body of the driverless vehicle, such as laser radar, image acquisition unit, ultrasonic sensor, millimeter wave radar and satellite positioning signal receiving antenna, etc. The laser radar is used to detect obstacles in front, and the image acquisition unit, i.e. the camera, is used to obtain environmental information around the vehicle when it is driving.

[0053] The test environment may include a rain test environment, a fog test environment, etc. to simulate the vehicle driving in extreme weather, and the test environment may also include an obstruction test environment to simulate the vehicle driving on a road with many buildings. It can be understood that the embodiment of the present invention does not limit the specific type of the test environment, and the complex driving environment that may affect the vehicle perception system is within the scope of the test environment of this embodiment.

[0054] Specifically, a certain test environment is set, such as a rain test environment, to obtain test collection data of the vehicle perception system in the rain test environment, for example, image information captured by an image acquisition unit in the rain test environment; or point cloud information collected by a lidar in the rain test environment, etc.

[0055] In addition, the test environment can be constructed artificially. By constructing different test environments artificially, various complex driving environments of vehicles can be simulated, thereby improving the adaptability of the test method provided by the embodiment of the present invention.

[0056] S120: Determine correction data according to the test collection data.

[0057] It is understandable that in complex driving environments, the information collected by the vehicle perception system may be abnormal, which may cause the onboard controller to make wrong instructions or fail to make the instructions that need to be completed. For example, in a rainy environment, raindrops may refract the detection laser of the lidar, resulting in an error in the judgment of the distance of obstacles or the inability to identify obstacles; in another example, in a rainy environment, due to the attachment of rain, the image collected by the camera may be distorted, thereby affecting the collection of surrounding environment information and thus affecting the calculation of the vehicle's driving route.

[0058] Therefore, the test data collected under different test environments are abnormal data that may affect vehicle driving, and the test data needs to be corrected to eliminate abnormal interference information in the test data to the greatest extent.

[0059] Specifically, after acquiring the test collection data of the vehicle perception system under different test environments, the test collection data collected by the vehicle perception system is exported, and the test collection data is processed to determine the correction data, which refers to the data used to correct the test collection data collected under different test environments. Exemplarily, if the test environment is a rain test environment, the image information collected by the image acquisition unit under the rain test environment can be exported. If the image information is found to be distorted, the image information can be repaired, spliced, etc. The data involved in the repair, splicing, etc. of the image information is the correction data.

[0060] S130: Establishing a corresponding relationship between the correction data and the test environment.

[0061] Specifically, after determining the correction data corresponding to the test collection data of each vehicle perception system under a certain test environment, the test scene, the test collection data corresponding to the test scene, and the correction data can be stored in the relevant vehicle perception system through model training and other methods. When the vehicle encounters a complex driving environment corresponding to a certain test environment during actual driving, each vehicle perception system can directly call the pre-stored correction data. The data collected by the current vehicle perception system is corrected by the correction data, which makes it easier for the on-board controller to obtain data corresponding to the actual situation and make correct judgments.

[0062] The technical solution in the embodiment of the present invention first obtains the test data collected by each vehicle perception system under different test environments, then corrects the test data, determines the correction data, and finally establishes the relationship between the correction data and the test environment. Through the above method, when the vehicle is driving in different complex environments, the vehicle perception system can directly call the correction data corresponding to this environment based on the collected environmental information data, thereby obtaining environmental information data with less interference, so that the on-board controller can make correct control instructions, avoiding driving accidents when the vehicle is driving in complex environments.

[0063] Based on the above embodiment, the present invention also provides another method for testing an unmanned vehicle. Figure 2 This is a flow chart of another unmanned vehicle testing method provided by an embodiment of the present invention. This embodiment is optimized based on the above embodiment. Figure 2 , the method of this embodiment specifically includes:

[0064] S210, obtaining test collection data of the vehicle perception system under different test environments.

[0065] S220, obtaining benchmark collection data of the vehicle perception system under a benchmark test environment; the benchmark test environment is a test environment without additional test conditions; and determining correction data based on the test collection data and the benchmark collection data.

[0066] Among them, the benchmark test environment refers to a test environment without external test conditions, corresponding to the vehicle driving under normal conditions. For example, if the test environment is a rain test environment, the benchmark test environment is a non-rain test environment.

[0067] Furthermore, the baseline collected data of the vehicle perception system under the benchmark test environment is obtained, and the baseline collected data is compared with the test collected data to calculate the correction data. It is understandable that the baseline collected data and the test collected data of the same vehicle perception system should be compared, and the calculated correction data is also the correction data of the vehicle perception system.

[0068] Exemplarily, if the test environment is a rain test environment and the vehicle perception system is a lidar, then the point cloud information collected by the lidar in the rain test environment can be obtained. In addition, the point cloud information collected by the lidar of the vehicle in a non-rain test environment is obtained, and the point cloud information in the rain test environment is stitched, filtered, and other processing is performed to make it close to the point cloud information collected by the lidar in the non-rain test environment or the same as the point cloud information collected by the lidar in the non-rain test environment. According to the above method, the correction data of the lidar in the rain test environment is determined.

[0069] S230: Establishing a corresponding relationship between the correction data and the test environment.

[0070] In this embodiment, the test collection data is corrected according to the benchmark test data, which can improve the accuracy of the correction data calculation and improve the efficiency of obtaining the correction data to a certain extent.

[0071] Optionally, in one embodiment, after establishing the corresponding relationship between the correction data and the test environment, the method may further include: correcting the test collection data according to the correction data. Figure 3 is a flowchart of another method for testing an unmanned vehicle provided by an embodiment of the present invention, with reference to Figure 3 , the method comprising:

[0072] S310, obtaining test collection data of the vehicle perception system under different test environments.

[0073] S320: Determine correction data according to the test collection data.

[0074] S330: Establish a corresponding relationship between the correction data and the test environment.

[0075] The specific implementation methods of the above steps are the same as those in the above embodiments and will not be repeated here.

[0076] S340, correcting the test collection data according to the correction data.

[0077] Specifically, after the correspondence between the correction data and the test environment is established, the test can be performed again under the same test environment. During this test, if test acquisition data corresponding to a certain correction data appears, the correction data can be applied to process the test acquisition data to correct the test acquisition data. For example, if the test environment is a rain test environment and the vehicle perception system is a laser radar, the point cloud information collected by the laser radar in the rain test environment can be filtered, spliced, combined, etc. according to the correction data to repair the collected point cloud information data.

[0078] In addition, if the test acquisition data still has abnormalities after the correction data is applied to correct the test acquisition data when the test is performed again under the same test environment, S310 to S330 can be repeated to update the correction data until the corrected test acquisition data has no abnormalities or is as close to the test acquisition data under the test environment without the additional test environment as possible.

[0079] The technical solution in the embodiment of the present invention repairs the test collection data according to the correction data to reduce the abnormal information data in the test collection data and improve the accuracy of the test collection data under different test environments.

[0080] Optionally, in one embodiment, before acquiring the test collection data of the vehicle perception system under different test environments, the method may further include: adjusting the test environment. Figure 4 is a flow chart of another unmanned vehicle testing method provided by an embodiment of the present invention. The technical solution in this embodiment is further refined on the basis of the above embodiment. Figure 4 , the method specifically comprises:

[0081] S410. Adjust the test environment.

[0082] It can be seen from the above embodiments that the test environment simulates a driving environment that may affect the vehicle perception system. In this embodiment, various external factors in the test environment and vehicle-specific factors can also be adjusted.

[0083] Optionally, the test environment may include a rain test environment;

[0084] Adjusting the test environment may include:

[0085] Adjust at least one of the pipeline pressure value, the rain time and the pipeline flow rate of the rain test equipment;

[0086] Adjust at least one of the pitch angle, roll angle and heading angle of the vehicle.

[0087] The test environment may include a rain test environment to simulate the vehicle driving in a rainy environment. This embodiment does not limit the construction method of the rain test environment. Preferably, a rain test room is built to collect test data in the rain test room. By building a rain test environment in the rain test room, water can be sprayed on the vehicle from multiple angles at the same time, restoring the real rain environment to the greatest extent.

[0088] Furthermore, the rain test environment can be adjusted by adjusting the pipeline pressure value, rain time and / or pipeline flow of the rain test equipment in the rain test room to obtain test collection data of the vehicle perception system under different rain intensities.

[0089] In addition, in the embodiment of the present invention, the vehicle body shape can also be adjusted during the test process to more realistically simulate the driving conditions of various vehicles in a rainy environment, for example: adjusting the pitch angle of the vehicle to simulate the vehicle driving on a slope; another example: adjusting the roll angle of the vehicle to simulate the vehicle tilting during driving; another example: adjusting the heading angle of the vehicle to simulate the vehicle turning.

[0090] It should be noted that when adjusting the rain test equipment or the shape of the vehicle itself in the above-mentioned rain test environment, the various parameters involved should be adjusted within a certain range. For example, when adjusting the various parameters of the rain test equipment, it should be ensured that the precipitation will not cause water to enter the vehicle; when adjusting the pitch angle of the vehicle, it can be adjusted within a slope slightly larger than the maximum slope of the existing road; when adjusting the roll angle of the vehicle, it should be adjusted within a range where the vehicle will not roll over.

[0091] S420, obtaining test collection data of the vehicle perception system under different rain test environments.

[0092] Specifically, different rain test environments can be built by changing the parameters of the rain test equipment or the vehicle body shape, and test collection data of the vehicle perception system under different rain test environments can be obtained. The test collection data is the same as in the above embodiment and will not be repeated here.

[0093] Optionally, the embodiment of the present invention preferably adjusts only one parameter of the rain test equipment or the vehicle body shape each time to ensure that more real rainfall environments are covered. For example, sudden rainstorms can be simulated by increasing the pipe pressure value; the precipitation can be controlled by adjusting the rain time to simulate rainfall environments with different rain intensities.

[0094] Acquiring test data collected by the vehicle perception system under different rain test environments can make the coverage of the test data wider and the test data closer to the information data collected by each perception system when the vehicle is actually driving in a complex environment.

[0095] S430: Determine correction data according to the test collection data.

[0096] S440. Establish a corresponding relationship between the correction data and different rain test environments.

[0097] The specific implementation methods of the above steps are the same as those in the above embodiments and will not be repeated here.

[0098] By acquiring test collection data of the vehicle perception system under different rain test environments, the test method provided in this embodiment can be made more adaptable and comprehensive in coverage, thereby ensuring that when the vehicle is driving in different complex environments, each vehicle perception system can correct the collected data information, thereby ensuring safe driving of the vehicle.

[0099] Optionally, in one embodiment, the vehicle perception system may include an image acquisition module, the image acquisition module including an image acquisition unit and a wiper;

[0100] Before obtaining test data of the vehicle perception system under different test environments, the following may also be included:

[0101] Adjust the working state of the wiper according to the test environment, and the working state includes the swing amplitude and the swing frequency.

[0102] Specifically, the vehicle perception system may include an image acquisition module, which mainly refers to a camera module, and the image acquisition module includes an image acquisition unit and a wiper, and the image acquisition unit mainly refers to a camera. It can be understood that in order to ensure that the vehicle can drive normally in a complex environment such as a rainy environment, a wiper will be configured for the camera to scrape off raindrops and ensure the accuracy of the collected image information.

[0103] It is understandable that the swing amplitude and the swing frequency of the wiper are inversely proportional. If the swing amplitude is large, the swing frequency will decrease; if the swing amplitude is small, the swing frequency will increase. In this embodiment, different test environments can be set up, and under different test environments, the wiper is controlled to work at a preset swing amplitude and swing frequency. Among them, the test environment mainly includes a rain test environment. The adjustment method of the test environment can refer to the above embodiment and will not be repeated here. At this time, the image information collected by the image acquisition unit when the wiper works at a preset swing amplitude and swing frequency under different test environments can be obtained. It is understandable that the swing amplitude and swing frequency of the wiper are different under different rain intensities. When adjusting the working state of the wiper, as many raindrops attached to the image acquisition unit as possible should be scraped off without affecting the continuity of the image collected by the image acquisition unit.

[0104] Furthermore, the image information collected by the image acquisition unit under different test environments is obtained, and the image information collected is exported to determine whether the image information is clear or close to the image information collected under the test environment without additional test conditions. If the image information collected under the test environment does not meet the above conditions, the working state of the wiper is reset, including adjusting the swing amplitude and swing frequency of the wiper, etc. Then, the above steps are re-executed until the image information collected by the image acquisition unit under each test environment is clear or close to the image information collected under the test environment without additional test conditions.

[0105] Furthermore, a correspondence between the above adjustment process and different test environments is established through methods such as model training, and the correspondence is stored in the image acquisition module.

[0106] In this embodiment, the working state of the wiper can be adjusted under different test environments to ensure the accuracy of the image information collected by the image acquisition unit under each test environment, so that during the actual driving of the vehicle, the wiper can operate in different driving environments to adapt to the working state of the driving environment and ensure driving safety.

[0107] Optionally, in one embodiment, the vehicle perception system may include an image acquisition module, the image acquisition module including a plurality of image acquisition units disposed around the body of the unmanned vehicle;

[0108] Obtain test data of vehicle perception systems under different test environments, including:

[0109] Acquire test acquisition data of multiple image acquisition units under different test environments;

[0110] Determine the correction data based on the test collection data, including:

[0111] Determine a plurality of correction data according to a plurality of test collection data;

[0112] Determining average correction data based on the plurality of correction data;

[0113] Establish the corresponding relationship between the correction data and the test environment, including:

[0114] Establish the correspondence between the average correction data and the test environment.

[0115] According to the above, an embodiment of the present invention further provides a method for testing an unmanned vehicle. Figure 5 : is a flow chart of another unmanned vehicle testing method provided by an embodiment of the present invention, which specifically includes:

[0116] S510: Acquire test acquisition data of multiple image acquisition units under different test environments.

[0117] S520, determining a plurality of correction data according to a plurality of test collection data; and determining average correction data according to the plurality of correction data.

[0118] S530: Establish a corresponding relationship between the average correction data and the test environment.

[0119] Specifically, the body of the unmanned vehicle may be provided with multiple image acquisition units, and the image acquisition module determines a three-dimensional image of the driving environment around the vehicle based on the image information acquired by the multiple image acquisition units. In the embodiment of the present invention, test acquisition data of multiple image acquisition units under different test environments may be obtained, multiple correction data may be determined based on the multiple test acquisition data, and then average correction data may be determined based on the multiple correction data; finally, a corresponding relationship between the average correction data and the test environment may be established.

[0120] Exemplarily, in a rain test environment, image information collected by each image acquisition unit under different rain test environments is obtained. After the rain test is completed, the above image information is derived. Due to the influence of raindrops, each image information will have a certain error, so the three-dimensional image integrated by multiple image information will have errors. At this time, multiple correction data are determined based on the multiple image information, and then the average correction data is determined based on the multiple correction data, that is, the three-dimensional image information integrated by the multiple image information is corrected to eliminate the abnormality of the three-dimensional image and obtain accurate three-dimensional image information; finally, the relationship between the average correction data and the rain test environment is established and stored in the image acquisition module. The method for determining the correction data can refer to the above embodiment and will not be repeated here.

[0121] In this embodiment, the test acquisition data of multiple image acquisition units can be corrected to obtain average corrected data, and then the relationship between the average corrected data and the test environment is established, further ensuring that when the vehicle is driving in a complex environment, the vehicle perception system can repair the anomalies in the collected information data and improve driving safety.

[0122] Based on the same concept, an embodiment of the present invention further provides a test device for an unmanned vehicle, which is used to execute the test method for an unmanned vehicle provided by any embodiment of the present invention. Figure 6 1 is a schematic diagram of the structure of a test device for an unmanned vehicle provided in an embodiment of the present invention. The test device has all the technical features and corresponding beneficial effects of the test method for an unmanned vehicle provided in any embodiment of the present invention.

[0123] See also Figure 6 , the test device comprises:

[0124] A test data acquisition module 100 is used to acquire test data of the vehicle perception system under different test environments;

[0125] The correction data determination module 200 is used to determine the correction data according to the test collection data;

[0126] The corresponding relationship establishing module 300 is used to establish the corresponding relationship between the correction data and the test environment.

[0127] Optionally, the correction data determination module includes a baseline acquisition data acquisition unit and a correction data determination unit; the baseline acquisition data acquisition unit obtains baseline acquisition data of the vehicle perception system under a baseline test environment; the benchmark test environment is a test environment without external test conditions; the correction data determination unit is used to determine the correction data based on the test acquisition data and the baseline acquisition data.

[0128] Optionally, the test device may further include a test acquisition data correction module, which is used to correct the test acquisition data according to the correction data after the corresponding relationship between the correction data and the test environment is established.

[0129] Optionally, the testing device may further include a testing environment adjustment module for adjusting the testing environment.

[0130] Optionally, the test environment includes a rain test environment;

[0131] The test environment adjustment module includes a first adjustment unit and a second adjustment unit; the first adjustment unit is used to adjust at least one of the pipeline pressure value, rain time and pipeline flow of the rain test equipment; the second adjustment unit is used to adjust at least one of the pitch angle, roll angle and heading angle of the vehicle.

[0132] Optionally, the vehicle perception system may include an image acquisition module, which includes an image acquisition unit and a wiper;

[0133] The test device provided in the embodiment of the present invention may also include a working state adjustment module, which is used to adjust the working state of the wiper according to the test environment before obtaining the test collection data of the vehicle perception system under different test environments. The working state includes the swing amplitude and the swing frequency.

[0134] Optionally, the vehicle perception system may include an image acquisition module, which includes a plurality of image acquisition units disposed around the body of the unmanned vehicle;

[0135] The test acquisition data acquisition module can be specifically used to acquire the test acquisition data of multiple image acquisition units under different test environments;

[0136] The correction data determination module can be specifically used to determine a plurality of correction data according to a plurality of test collection data, and determine average correction data according to the plurality of correction data;

[0137] The corresponding relationship establishing module can be specifically used to establish the corresponding relationship between the average correction data and the test environment.

[0138] The test device for unmanned vehicles provided in the embodiment of the present invention first obtains the test data collected by each vehicle perception system under different test environments, then corrects the test data, determines the correction data, and finally establishes the relationship between the correction data and the test environment. Through the above method, when the vehicle is driving in different complex environments, the vehicle perception system can directly call the correction data corresponding to the environment based on the collected environmental information data, thereby obtaining environmental information data with less interference, so that the on-board controller can make correct control instructions, avoiding driving accidents when the vehicle is driving in complex environments.

[0139] Based on the same concept, an embodiment of the present invention also provides an unmanned vehicle, including one or more processors;

[0140] A storage device for storing one or more programs;

[0141] When one or more programs are executed by one or more processors, the one or more processors implement the unmanned vehicle testing method provided by any embodiment of the present invention.

[0142] In the unmanned vehicle in the embodiment of the present invention, the vehicle perception system can directly call the correction data corresponding to the environment based on the collected environmental information data, thereby obtaining environmental information data with less interference, so that the on-board controller can make correct control instructions and avoid driving accidents when the vehicle is driving in a complex environment.

[0143] An embodiment of the present invention further provides a medium on which a computer program is stored. When the program is executed by a computer processor, the test method for an unmanned vehicle provided by any embodiment of the present invention is implemented.

[0144] Through the above description of the implementation mode, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation mode. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of various embodiments of the present invention.

[0145] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for testing an unmanned vehicle, characterized in that: include: Obtain test data of vehicle perception systems under different test environments; Determining correction data according to the test acquisition data; Establishing a corresponding relationship between the correction data and the test environment; The vehicle perception system includes an image acquisition module, and the image acquisition module includes a plurality of image acquisition units arranged around the body of the unmanned vehicle; Obtain test data of vehicle perception systems under different test environments, including: Acquire test acquisition data of a plurality of image acquisition units under different test environments; Determining correction data according to the test collection data includes: Determine a plurality of the correction data according to a plurality of the test collection data; Determining average correction data based on the plurality of correction data; Establishing a correspondence between the correction data and the test environment includes: Establishing a corresponding relationship between the average correction data and the test environment; The correction data refers to data for correcting the test data collected under different test environments; After the corresponding relationship between the correction data and the test environment is established, the method further includes: Perform the test under the test environment again; If test collection data corresponding to any of the correction data appears, the test collection data is corrected by applying the correction data; If the test collection data still has abnormalities after the correction data is applied to correct the test collection data, the correction data is updated until the corrected test collection data has no abnormalities or is as close as possible to the test collection data in the absence of an additional test environment.

2. The testing method according to claim 1, characterized in that: Determining correction data according to the test collection data includes: Obtaining benchmark collection data of the vehicle perception system in a benchmark test environment; the benchmark test environment is a test environment without additional test conditions; Correction data is determined based on the test acquisition data and the reference acquisition data.

3. The testing method according to claim 1, characterized in that: After establishing the correspondence between the correction data and the test environment, the method further includes: The test collection data is corrected according to the correction data.

4. The testing method according to claim 1, characterized in that: Before obtaining the test data of the vehicle perception system under different test environments, it also includes: Adjust the test environment.

5. The testing method according to claim 4, characterized in that: The test environment includes a rain test environment; Adjusting the test environment includes: Adjust at least one of the pipeline pressure value, the rain time and the pipeline flow rate of the rain test equipment; Adjust at least one of the pitch angle, roll angle and heading angle of the vehicle.

6. The testing method according to claim 1, characterized in that: The vehicle perception system includes an image acquisition module, and the image acquisition module includes an image acquisition unit and a wiper; Before obtaining the test data of the vehicle perception system under different test environments, it also includes: The working state of the wiper is adjusted according to the test environment, and the working state includes the swing amplitude and the swing frequency.

7. A test device for an unmanned vehicle, characterized in that: include: A test data acquisition module is used to acquire test data of the vehicle perception system under different test environments; A correction data determination module, used to determine the correction data according to the test collection data; A corresponding relationship establishing module, used to establish a corresponding relationship between the correction data and the test environment; The vehicle perception system includes an image acquisition module, and the image acquisition module includes a plurality of image acquisition units arranged around the body of the unmanned vehicle; The test acquisition data acquisition module is also used to acquire the test acquisition data of multiple image acquisition units under different test environments; The correction data determination module is further used to determine a plurality of the correction data according to the plurality of the test collection data, and to determine average correction data according to the plurality of the correction data; The corresponding relationship establishing module is also used to establish the corresponding relationship between the average correction data and the test environment; The correction data refers to data for correcting the test data collected under different test environments; The corresponding relationship verification module is used to perform the test under the test environment again; if test collection data corresponding to any of the correction data appears, the correction data is applied to correct the test collection data; if the test collection data is still abnormal after the correction data is applied to correct the test collection data, the correction data is updated until the corrected test collection data has no abnormality or is as close as possible to the test collection data under the test environment without the additional test.

8. An unmanned vehicle, characterized in that: The unmanned vehicle comprises: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the testing method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the testing method according to any one of claims 1 to 6 is implemented.

Images