Pedestrian-vehicle-road cloud intelligent networking algorithm test system

Abstract

The invention discloses a human-vehicle-road cloud intelligent networking algorithm test system. Data transmission and processing exist among the road data module, the real-time system module, the driving simulation module and the state acquisition module in the man-vehicle-road cloud intelligent network connection algorithm test system, the four modules form a complete test, and the overall performance of the algorithm is tested not only in the last real vehicle test stage. Therefore, the function and performance of the algorithm can be completely tested in a laboratory, real-time road scene data, road vehicle data, vehicle dynamic simulation and driving feeling information acquisition are integrated in the test process, scene-in-the-loop, hardware-in-the-loop and driver and passenger-in-the-loop are integrated, and the algorithm test does not stay on the data level any more. The finally developed algorithm can meet the real vehicle condition, the algorithm test can be carried out through the driving simulation module, the test system can quickly complete the test verification of the algorithm function and performance in a laboratory, and the development progress of the intelligent network connection regulation and control algorithm is accelerated.

Description

Technical Field

[0001] The present invention relates to the field of intelligent network connection technology, and in particular to a human-vehicle-road-cloud intelligent network connection algorithm testing system. Background Art

[0002] Currently, the testing method for intelligent connected algorithms in the R&D and verification stages is mainly simulation testing, and the four stages of testing intelligent connected algorithms are completed independently, with each stage having corresponding testing methods and evaluation content.

[0003] For example: model-in-the-loop testing in the model development phase, software-in-the-loop testing after code generation, hardware-in-the-loop testing after software and hardware are combined, and actual vehicle testing; the evaluation contents corresponding to the testing methods in different stages are also different. The test object of the model-in-the-loop testing is the control algorithm model, and the evaluation content is aimed at the specific logical jumps of the model; the software-in-the-loop testing is aimed at whether the function of the model is consistent with the model after the code is generated, and the evaluation content is mainly aimed at whether the output result of the code is consistent with the logic of the model; the hardware-in-the-loop testing is mainly aimed at whether the logic after the control algorithm and hardware are combined is consistent with the functional requirements; the actual vehicle testing is mainly aimed at whether the overall performance of the algorithm is consistent with the functional requirements.

[0004] It can be seen that the testing methods and evaluation contents of each stage are different, resulting in the separation of each stage from each other. In the three stages before the actual vehicle test, only the functional logic can be tested, but the performance of the algorithm cannot be tested, nor can the performance under actual road scenarios and data be tested. As a result, the testing of the algorithm only stays at the data level, resulting in the final developed algorithm may not be able to adapt to the actual situation. The overall performance of the algorithm can only be measured in the actual vehicle test, resulting in the inability to fully verify the function and performance of the algorithm. The actual vehicle test can only be carried out after the completion of the first three stages, resulting in the inability to quickly verify the function and performance of the algorithm. Summary of the invention

[0005] The present invention provides a human-vehicle-road-cloud intelligent network algorithm testing system, which can quickly and completely verify the function and performance of the algorithm. The specific technical solution is as follows.

[0006] In a first aspect, the present invention provides a human-vehicle-road-cloud intelligent network algorithm testing system, comprising: a road data module, a real-time system module, a driving simulation module and a state acquisition module;

[0007] The road data module collects the road condition data of the current road in real time, filters out the road scene data and the road vehicle data from the road condition data, and sends them to the real-time system module;

[0008] The real-time system module tests the intelligent network-connected regulation and control algorithm to be tested based on the road scene data, the road vehicle data, the vehicle interface data of the virtual vehicle to be tested, and the vehicle dynamics model of the virtual vehicle to be tested, obtains the motion control information of the virtual vehicle to be tested and sends it to the driving simulation module, wherein the intelligent network-connected regulation and control algorithm to be tested is used to perform lateral, longitudinal and vertical motion control on the virtual vehicle to obtain the motion control information of the virtual vehicle;

[0009] The driving simulation module performs vehicle dynamic simulation based on the motion control information;

[0010] The state acquisition module collects driving experience information of the driver and passengers when the driving simulation module performs vehicle dynamic simulation, evaluates the intelligent network connection regulation and control algorithm to be tested based on the driving experience information to obtain a first evaluation result, evaluates the intelligent network connection regulation and control algorithm to be tested based on the motion control information to obtain a second evaluation result, and obtains a final evaluation result based on the first evaluation result and the second evaluation result.

[0011] Optionally, the road condition data includes at least road scene data of the current road, road vehicle data, road construction information, road accident information and road congestion information.

[0012] Optionally, the real-time system module includes a host computer and a real-time machine that are communicatively connected to each other, and the host computer is provided with test management software, automated test software and a vehicle dynamics model of the virtual vehicle to be tested;

[0013] The test management software deploys the automated test software and the vehicle dynamics model of the virtual vehicle to be tested to the real-time machine before testing;

[0014] The real-time machine tests the intelligent connected regulation and control algorithm to be tested based on the road scene data, the road vehicle data, the whole vehicle interface data of the virtual vehicle to be tested, and the vehicle dynamics model of the virtual vehicle to be tested through the automated testing software.

[0015] Optionally, the real-time system module further comprises a communication module communicatively connected to the real-time machine;

[0016] The real-time machine sends the motion control information of the virtual vehicle to be tested to the driving simulation module through the communication module.

[0017] Optionally, the real-time system module further includes a data simulation module communicatively connected to the communication module;

[0018] The data simulation module configures the vehicle interface data of the virtual vehicle to be tested and sends it to the real-time machine through the communication module.

[0019] Optionally, the motion control information includes at least lateral motion control information, longitudinal motion control information and vertical motion control information of the vehicle.

[0020] Optionally, the state acquisition module includes a camera, a wearable device and a data analysis unit;

[0021] The camera collects the facial expression information and eye state information of the driver and passenger during the vehicle dynamic simulation by the driving simulation module and sends the information to the data analysis unit;

[0022] The wearable device collects blood oxygen concentration information of a target area of ​​the cerebral cortex of a driver or passenger during a vehicle dynamic simulation by the driving simulation module and sends the information to the data analysis unit;

[0023] The data analysis unit evaluates the intelligent network-connected regulation and control algorithm to be tested based on the facial expression information, the eye state information and the blood oxygen concentration information to obtain a first evaluation result.

[0024] Optionally, the data analysis unit determines whether the driver and passenger are in one of the following states: fatigue, tension, panic and excitement based on the facial expression information, the eye state information and the blood oxygen concentration information; if not, it determines that the first evaluation result is that the intelligent network-connected regulation and control algorithm to be tested has reasonable motion control over the virtual vehicle; if so, it determines that the first evaluation result is that the intelligent network-connected regulation and control algorithm to be tested has unreasonable motion control over the virtual vehicle.

[0025] Optionally, the data analysis unit determines whether the motion control information is motion control information for normal driving. If so, it determines that the second evaluation result is that the motion control of the virtual vehicle by the intelligent networked regulation and control algorithm to be tested is reasonable. If not, it determines that the second evaluation result is that the motion control of the virtual vehicle by the intelligent networked regulation and control algorithm to be tested is unreasonable.

[0026] Optionally, when both the first evaluation result and the second evaluation result indicate that the motion control of the virtual vehicle by the intelligent connected control algorithm to be tested is reasonable, the data analysis unit determines that the final evaluation result is passed.

[0027] From the above content, it can be seen that an embodiment of the present invention provides a human-vehicle-road-cloud intelligent network algorithm testing system, including a road data module, a real-time system module, a driving simulation module and a state acquisition module. The road data module collects the road condition data of the current road in real time, filters out road scene data and road vehicle data from the road condition data, and sends them to the real-time system module; the real-time system module tests the intelligent network control algorithm to be tested based on the road scene data, road vehicle data, the whole vehicle interface data of the virtual vehicle to be tested, and the vehicle dynamics model of the virtual vehicle to be tested, obtains the motion control information of the virtual vehicle to be tested and sends it to the driving simulation module, wherein the intelligent network control algorithm to be tested is used to perform lateral, longitudinal and vertical motion control on the virtual vehicle to obtain the motion control information of the virtual vehicle. The driving simulation module performs vehicle dynamic simulation based on motion control information; the state acquisition module collects driving experience information of the driver and passengers during the vehicle dynamic simulation process of the driving simulation module, evaluates the intelligent network connection regulation and control algorithm to be tested based on the driving experience information to obtain a first evaluation result, evaluates the intelligent network connection regulation and control algorithm to be tested based on the motion control information to obtain a second evaluation result, and obtains a final evaluation result based on the first evaluation result and the second evaluation result. Therefore, in the embodiment of the present invention, data transmission and processing exist between the four modules, namely, the road data module, the real-time system module, the driving simulation module and the status acquisition module, and they are connected to each other. Moreover, the four modules form a complete test, and the overall performance of the algorithm can be measured not only in the final actual vehicle test stage. Therefore, the function and performance of the algorithm can be fully tested and verified in the laboratory. In the process of intelligent network algorithm testing, real-time road scene data, road vehicle data, vehicle dynamic simulation and driving experience information collection are integrated, and the scene in the loop, hardware in the loop, and driver and passenger in the loop are integrated, so that the test of the algorithm not only stays at the data level, but also enables the final developed algorithm to meet the actual vehicle situation. Moreover, the test of the algorithm can be carried out through the driving simulation module without the need for actual vehicle testing. The test verification of the algorithm function and performance can be quickly completed in the laboratory, thereby accelerating the development progress of the intelligent network regulation and control algorithm.

[0028] The innovative features of the embodiments of the present invention include:

[0029] 1. In the embodiment of the present invention, data transmission and processing exist between the four modules, namely, the road data module, the real-time system module, the driving simulation module and the status acquisition module. The modules are interconnected and the four modules form a complete test. The overall performance of the algorithm can be measured not only in the final actual vehicle test stage. Therefore, the function and performance of the algorithm can be fully tested in the laboratory. In the process of intelligent network algorithm testing, real-time road scene data, road vehicle data, vehicle dynamic simulation and driving experience information collection are integrated, and the scene in the loop, hardware in the loop and driver in the loop are integrated, so that the algorithm test not only stays at the data level, but also enables the final developed algorithm to meet the actual vehicle situation. Moreover, the algorithm test can be carried out through the driving simulation module without the need for actual vehicle testing. The algorithm function and performance test verification can be quickly completed in the laboratory, thereby accelerating the development progress of the intelligent network control algorithm.

[0030] 2. Real-time road scene data, road vehicle data, vehicle dynamic simulation and driving experience information collection are integrated into the test, integrating the scene in the loop, hardware in the loop, and driver and passenger in the loop, making it possible to automatically evaluate the driver and passenger's feelings about the intelligent network algorithm functions during the algorithm development stage.

[0031] 3. Before testing, the automated testing software and the vehicle dynamics model of the virtual vehicle to be tested are deployed to the real-time machine through the test management software. Then, the intelligent connected regulation and control algorithm to be tested is tested through the automated testing software based on the road scene data, road vehicle data, vehicle interface data of the virtual vehicle to be tested, and the vehicle dynamics model of the virtual vehicle to be tested, thereby achieving fully automatic testing.

[0032] 4. Configure the vehicle interface data of the virtual vehicle to be tested through the data simulation module.

[0033] 5. By setting up cameras and wearable devices, the facial expression information, eye state information and blood oxygen concentration information of the driver and passengers during the vehicle dynamic simulation process can be obtained, so that the intelligent network control algorithm to be tested can be evaluated based on the facial expression information, eye state information and blood oxygen concentration information to obtain the first evaluation result, thereby realizing the evaluation of the algorithm based on the feelings of the driver and passengers.

[0034] 6. The data analysis unit determines whether the driver or passenger is in one of the following states: fatigue, tension, panic, or excitement through facial expression information, eye state information, and blood oxygen concentration information. If not, the first evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent network connection control algorithm to be tested is reasonable. Otherwise, the first evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent network connection control algorithm to be tested is unreasonable. This achieves the purpose of automatically evaluating the driver or passenger's feelings about the intelligent network connection algorithm function during the algorithm development stage.

[0035] 7. The data analysis unit determines whether the motion control information is the motion control information of normal driving. If yes, it means that the motion control information meets the requirements of normal driving, and determines that the second evaluation result is that the motion control of the virtual vehicle by the intelligent network control algorithm to be tested is reasonable. If no, it determines that the second evaluation result is that the motion control of the virtual vehicle by the intelligent network control algorithm to be tested is unreasonable. In this way, the intelligent network control algorithm to be tested can be evaluated at the data level.

[0036] 8. When both the first and second evaluation results show that the motion control of the virtual vehicle by the intelligent network control algorithm to be tested is reasonable, the data analysis unit determines that the final evaluation result is passed. The intelligent network control algorithm to be tested is evaluated at both the human perception level and the data level, which improves the accuracy of the evaluation and ensures that the final developed algorithm can meet the actual vehicle effect.

[0037] Of course, it is not necessary to achieve all of the advantages described above at the same time to implement any product or method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative work.

[0039] Figure 1 A structural schematic diagram of a human-vehicle-road-cloud intelligent network algorithm testing system provided by an embodiment of the present invention;

[0040] Figure 2 Another structural schematic diagram of the human-vehicle-road-cloud intelligent network algorithm testing system provided in an embodiment of the present invention.

[0041] Figure 1-Figure 2Among them, 1 road data module, 11 cloud brain, 2 real-time system module, 21 host computer, 22 real-time machine, 23 communication module, 24 data simulation module, 3 driving simulation module, 4 state acquisition module, 5 intelligent driving algorithm controller. DETAILED DESCRIPTION

[0042] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] It should be noted that the terms "including" and "having" and any variations thereof in the embodiments of the present invention and the accompanying drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0044] The embodiment of the present invention discloses a human-vehicle-road-cloud intelligent network algorithm testing system, which can quickly and completely verify the function and performance of the algorithm. The embodiment of the present invention is described in detail below.

[0045] Figure 1 A structural schematic diagram of a human-vehicle-road-cloud intelligent network algorithm testing system provided in an embodiment of the present invention. Figure 2 Another structural diagram of the human-vehicle-road-cloud intelligent network algorithm testing system provided by the embodiment of the present invention. Figure 1 The human-vehicle-road-cloud intelligent network algorithm testing system provided in an embodiment of the present invention includes a road data module 1, a real-time system module 2, a driving simulation module 3 and a state acquisition module 4.

[0046] Among them, the road data module 1 is used to obtain road scene data and road vehicle data; the real-time system module 2 is used to test the intelligent network control algorithm; the driving simulation module 3 is used to simulate the vehicle dynamics of the vehicle equipped with the intelligent network function, specifically, to simulate the dynamic performance of the vehicle equipped with the intelligent network function in six degrees of freedom (i.e. forward, lateral, vertical, pitch, roll, and yaw); the state acquisition module 4 is used to obtain the driving experience information of the driver and passengers when the intelligent network function is turned on, and to evaluate the intelligent network control algorithm. The intelligent network algorithm in the title is the intelligent network control algorithm.

[0047] Specifically, the road data module 1 collects the road condition data of the current road in real time, filters out the road scene data and the road vehicle data from the road condition data, and sends them to the real-time system module 2 .

[0048] In order to test the intelligent network algorithm of human, vehicle, road and cloud, it is necessary to set up a road data module 1 on the road in the intelligent network demonstration area, where Figure 2 The road data module 1 may include: RSU (Road Side Unit, intelligent roadside equipment), roadside camera, roadside radar, MEC (Mobility Edge Computing, roadside computing unit) and cloud brain 11. The cloud brain is the cloud control platform. Figure 2 In order to facilitate viewing, Cloud Brain 11 is drawn separately.

[0049] The camera and radar collect the traffic data of the current road and send it to the roadside computing unit MEC. MEC integrates and calculates the traffic data and sends it to the cloud brain 11 through the RSU. The cloud brain 11 filters out the road scene data and road vehicle data from the traffic data through the road-cloud protocol and sends them to the real-time system module 2.

[0050] Among them, the road condition data at least includes the road scene data, road vehicle data, road construction information, road accident information and road congestion information of the current road; the road scene data at least includes the lane line information, road sign information and traffic light information of the current road; the road vehicle data at least includes the number of vehicles on the current road, the vehicle's driving direction information and the driving speed information.

[0051] The real-time system module 2 receives the road scene data and road vehicle data sent by the road data module 1, and tests the intelligent network-connected regulation and control algorithm to be tested based on the road scene data, road vehicle data, the whole vehicle interface data of the virtual vehicle to be tested, and the vehicle dynamics model of the virtual vehicle to be tested, and obtains the motion control information of the virtual vehicle to be tested and sends it to the driving simulation module 3, wherein the intelligent network-connected regulation and control algorithm to be tested is used to perform lateral, longitudinal and vertical motion control on the virtual vehicle to obtain the motion control information of the virtual vehicle, and the motion control information at least includes the lateral motion control information, longitudinal motion control information and vertical motion control information of the vehicle.

[0052] The driving simulation module 3 receives the motion control information sent by the real-time system module 2, and performs vehicle dynamic simulation based on the motion control information. Specifically, the driving simulation module 3 may be a dynamic driving simulator.

[0053] The state acquisition module 4 collects the driving experience information of the driver and passengers when the driving simulation module 3 is performing vehicle dynamic simulation, and evaluates the intelligent network connection regulation and control algorithm to be tested based on the driving experience information to obtain a first evaluation result, and evaluates the intelligent network connection regulation and control algorithm to be tested based on the motion control information to obtain a second evaluation result, and obtains a final evaluation result based on the first evaluation result and the second evaluation result.

[0054] From the above content, it can be seen that the human-vehicle-road-cloud intelligent network algorithm testing system provided in this embodiment includes a road data module 1, a real-time system module 2, a driving simulation module 3 and a state acquisition module 4. The road data module 1 collects the road condition data of the current road in real time, filters out the road scene data and road vehicle data from the road condition data, and sends them to the real-time system module; the real-time system module 2 tests the intelligent network control algorithm to be tested based on the road scene data, road vehicle data, the whole vehicle interface data of the virtual vehicle to be tested, and the vehicle dynamics model of the virtual vehicle to be tested, obtains the motion control information of the virtual vehicle to be tested and sends it to the driving simulation module, wherein the intelligent network control algorithm to be tested is used to perform lateral, longitudinal and vertical motion control on the virtual vehicle to obtain the motion control information of the virtual vehicle. The driving simulation module 3 performs vehicle dynamic simulation based on the motion control information; the state acquisition module 4 collects the driving feeling information of the driver and passengers during the vehicle dynamic simulation process of the driving simulation module, and evaluates the intelligent network connection regulation and control algorithm to be tested based on the driving feeling information to obtain a first evaluation result, and evaluates the intelligent network connection regulation and control algorithm to be tested based on the motion control information to obtain a second evaluation result, and obtains a final evaluation result based on the first evaluation result and the second evaluation result. Therefore, in the embodiment of the present invention, data transmission and processing exist between the four modules of the road data module 1, the real-time system module 2, the driving simulation module 3 and the status acquisition module 4, and they are connected with each other. Moreover, the four modules form a complete test, and the overall performance of the algorithm can be measured not only in the final actual vehicle test stage. Therefore, the function and performance of the algorithm can be fully tested and verified in the laboratory. In the process of intelligent network algorithm testing, real-time road scene data, road vehicle data, vehicle dynamic simulation and driving experience information collection are integrated, and the scene in the loop, hardware in the loop, and driver and passenger in the loop are integrated, so that the test of the algorithm is no longer limited to the data level, so that the algorithm finally developed can meet the actual vehicle situation, and the algorithm can be tested through the driving simulation module without the need for actual vehicle testing. The algorithm function and performance test verification can be quickly completed in the laboratory, thereby accelerating the development progress of the intelligent network control algorithm.

[0055] In addition, real-time road scene data, road vehicle data, vehicle dynamic simulation and driving experience information collection are incorporated into the test, integrating the scene, hardware and driver and passenger in the loop, and achieving automated evaluation of the driver and passenger's feelings about the intelligent connected algorithm functions during the algorithm development stage.

[0056] Continue to see Figure 2 In one implementation, the real-time system module 2 may include a host computer 21 and a real-time machine 22 that are communicatively connected to each other, wherein the host computer 21 is provided with test management software, automated test software, and a vehicle dynamics model of a virtual vehicle to be tested. Figure 2 The arrows in the figure indicate the direction of information transmission.

[0057] Among them, the host computer 21 is used to run the software and the operation interface; the real-time machine 22 is used to perform high-performance computing. The test management software is used to configure the interface, and also deploy the model of the host computer 21 to the real-time machine; the automated test software is used to traverse the test cases and automatically execute the test; the vehicle dynamics model of the virtual vehicle to be tested can be configured according to the needs of the actual test vehicle. For example: if the intelligent network control algorithm configured on the car is developed, the vehicle dynamics model of the virtual vehicle to be tested is the vehicle dynamics model of the car, and the configuration method of the vehicle dynamics model of the virtual vehicle to be tested is any vehicle dynamics model configuration method in the prior art, and the embodiment of the present invention does not make any limitation to this.

[0058] The test management software deploys the automated test software and the vehicle dynamics model of the virtual vehicle to be tested to the real-time machine 22 before the test is performed.

[0059] The real-time machine 22 tests the intelligent network control algorithm to be tested based on the road scene data, road vehicle data, vehicle interface data of the virtual vehicle to be tested, and the vehicle dynamics model of the virtual vehicle to be tested through the automated testing software. In one implementation, the intelligent network control algorithm to be tested is set in the real-time machine 22, and in another implementation, the intelligent network control algorithm to be tested is set in the intelligent driving algorithm controller 5. Figure 2 shown.

[0060] Therefore, the automated testing software and the vehicle dynamics model of the virtual vehicle to be tested are deployed to the real-time machine through the test management software before the test, and then the intelligent connected regulation and control algorithm to be tested is tested through the automated testing software based on the road scene data, road vehicle data, the vehicle interface data of the virtual vehicle to be tested, and the vehicle dynamics model of the virtual vehicle to be tested, thereby realizing fully automatic testing.

[0061] Continuing to refer to 2, in one implementation, when the real-time system module 2 includes a host computer 21 and a real-time machine 22 that are communicatively connected to each other, the real-time system module 2 may also include a communication module 23 that is communicatively connected to the real-time machine 22, wherein the communication module 23 is used for transmitting information.

[0062] The real-time machine 22 sends the motion control information of the virtual vehicle to be tested to the driving simulation module 3 through the communication module 23 .

[0063] In an implementation, when the real-time system module 2 further includes a communication module 23 , the real-time system module 2 further includes a data simulation module 24 communicatively connected to the communication module 23 .

[0064] The data simulation module 24 configures the vehicle interface data of the virtual vehicle to be tested and sends it to the real-time machine 22 through the communication module 23 , wherein the vehicle interface data may at least include sensor data, operating voltage and operating current of the vehicle.

[0065] Therefore, the whole vehicle interface data of the virtual vehicle to be tested is configured through the data simulation module 24 .

[0066] In one implementation, the state acquisition module 4 includes a camera, a wearable device, and a data analysis unit, wherein the camera is installed around the driving simulation module 3, and the acquisition range includes at least the face of the driver and the passenger, the driver and the passenger wear the wearable device, and the camera and the wearable device are both connected to the data analysis unit in communication, and the wearable device can be a smart bracelet or a smart watch. The driver and the passenger refer to the driver and / or the passenger.

[0067] The camera collects the facial expression information and eye state information of the driver and passengers during the vehicle dynamic simulation process of the driving simulation module 3 and sends it to the data analysis unit, wherein the facial expression information may at least include concentration, excitement, fatigue, tension and panic, and the eye state information may at least include concentration, excitement, fatigue, tension and panic.

[0068] The wearable device collects the blood oxygen concentration information of the target area of ​​the cerebral cortex of the driver and passenger during the vehicle dynamic simulation by the driving simulation module 3 and sends it to the data analysis unit.

[0069] The data analysis unit then evaluates the intelligent connected regulation and control algorithm to be tested based on the facial expression information, eye state information and blood oxygen concentration information to obtain the first evaluation result.

[0070] Therefore, by setting up cameras and wearable devices, the facial expression information, eye state information and blood oxygen concentration information of the driver and passengers during the vehicle dynamic simulation process can be obtained, so that the intelligent connected regulation and control algorithm to be tested can be evaluated based on the facial expression information, eye state information and blood oxygen concentration information to obtain the first evaluation result, thereby realizing the evaluation of the algorithm based on the feelings of the driver and passengers.

[0071] In one implementation, when the state acquisition module 4 includes a camera, a wearable device and a data analysis unit, the data analysis unit determines whether the driver or passenger is in one of the following states: fatigue, tension, panic and excitement based on facial expression information, eye state information and blood oxygen concentration information. If not, the first evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent networked regulation and control algorithm to be tested is reasonable. If so, the first evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent networked regulation and control algorithm to be tested is unreasonable.

[0072] The data analysis unit may determine whether the driver or passenger is in one of a fatigued state, a nervous state, a panicked state, and an excited state based on the facial expression information, the eye state information, and the blood oxygen concentration information:

[0073] The data analysis unit determines whether the facial expression information and the eye state information are both tired, nervous, excited, or frightened, and determines whether the blood oxygen concentration information is within the blood oxygen concentration range of the cerebral cortex target area, wherein the cerebral cortex target area may include at least a left brain area and a right brain area;

[0074] If the facial expression information and eye state information of the driver and passenger are both tired and the blood oxygen concentration information is not within the blood oxygen concentration range of the target area of ​​the cerebral cortex, it is determined that the driver and passenger are in a tired state;

[0075] If the facial expression information and eye state information of the driver and passenger are both nervous and the blood oxygen concentration information is not within the blood oxygen concentration range of the target area of ​​the cerebral cortex, it is determined that the driver and passenger are in a nervous state;

[0076] If the facial expression information and eye state information of the driver and passenger are both panic and the blood oxygen concentration information is not within the blood oxygen concentration range of the target area of ​​the cerebral cortex, it is determined that the driver and passenger is in a panic state;

[0077] If the facial expression information and eye state information of the driver and passenger are both excited and the blood oxygen concentration information is not within the blood oxygen concentration range of the target area of ​​the cerebral cortex, it is determined that the driver and passenger are in an excited state.

[0078] If the driver and passenger are not in one of the states of fatigue, tension, panic and excitement, for example, they are in a focused state, and the blood oxygen concentration information is within the blood oxygen concentration range of the target area of ​​the cerebral cortex, it means that the driver and passenger are in a stable mood, which further indicates that the vehicle dynamic simulation performed by the driving simulation module 3 based on the motion control information is a motion simulation of normal driving. It can be determined that the first evaluation result is that the motion control of the virtual vehicle by the intelligent network control algorithm to be tested is reasonable. Otherwise, it is determined that the first evaluation result is that the motion control of the virtual vehicle by the intelligent network control algorithm to be tested is unreasonable.

[0079] If there are multiple drivers and passengers, the status of each driver and passenger can be judged. The judgment method is as described above and will not be repeated here. If the number of drivers and passengers who are not in one of the states of fatigue, tension, panic and excitement is greater than 1 / 2 of the total number of drivers and passengers, it can be determined that the first evaluation result is that the intelligent network connection regulation and control algorithm to be tested has reasonable motion control over the virtual vehicle. Otherwise, it is determined that the first evaluation result is that the intelligent network connection regulation and control algorithm to be tested has unreasonable motion control over the virtual vehicle.

[0080] Thus, the data analysis unit determines whether the driver is in one of the following states: fatigue, tension, panic, and excitement through facial expression information, eye state information, and blood oxygen concentration information. If not, the first evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent network connection regulation and control algorithm to be tested is reasonable. Otherwise, the first evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent network connection regulation and control algorithm to be tested is unreasonable. This achieves the purpose of automatically evaluating the driver's and passenger's feelings about the intelligent network connection algorithm function during the algorithm development stage.

[0081] In one implementation, the data analysis unit determines whether the motion control information is motion control information for normal driving. If so, the second evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent network control algorithm to be tested is reasonable. If not, the second evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent network control algorithm to be tested is unreasonable.

[0082] Generally speaking, if the vehicle is in a normal driving state, the motion control information meets the requirements of normal driving, for example, the vehicle speed cannot exceed the set vehicle speed threshold, etc. Therefore, when performing the evaluation, the data analysis unit can determine whether the motion control information is the motion control information of normal driving. If so, it means that the motion control information meets the requirements of normal driving, and the second evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent network connection regulation and control algorithm to be tested is reasonable. If not, the second evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent network connection regulation and control algorithm to be tested is unreasonable.

[0083] Thus, the data analysis unit determines whether the motion control information is the motion control information of normal driving. If yes, it means that the motion control information meets the requirements of normal driving, and determines that the second evaluation result is that the motion control of the virtual vehicle by the intelligent network connection regulation and control algorithm to be tested is reasonable. If no, it determines that the second evaluation result is that the motion control of the virtual vehicle by the intelligent network connection regulation and control algorithm to be tested is unreasonable. In this way, the intelligent network connection regulation and control algorithm to be tested can be evaluated at the data level.

[0084] In one implementation, the data analysis unit determines that the final evaluation result is passed when the first evaluation result and the second evaluation result both indicate that the motion control of the virtual vehicle by the intelligent connected regulation and control algorithm to be tested is reasonable.

[0085] If both the first evaluation result and the second evaluation result show that the motion control of the virtual vehicle by the intelligent network control algorithm to be tested is reasonable, it means that the motion control of the virtual vehicle by the intelligent network control algorithm to be tested is reasonable both in terms of human perception and data, and therefore, the final evaluation result can be determined as passing the evaluation. Otherwise, the final evaluation result is determined as failing the evaluation.

[0086] Therefore, when both the first evaluation result and the second evaluation result indicate that the motion control of the virtual vehicle by the intelligent network connection control algorithm to be tested is reasonable, the data analysis unit determines that the final evaluation result is passed. The intelligent network connection control algorithm to be tested is evaluated at both the human perception level and the data level, which improves the accuracy of the evaluation and ensures that the final developed algorithm can meet the actual vehicle effect.

[0087] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0088] Those skilled in the art can understand that the modules in the device in the embodiment can be distributed in the device in the embodiment according to the description of the embodiment, or can be changed accordingly and located in one or more devices different from the embodiment. The modules in the above embodiment can be combined into one module, or can be further divided into multiple sub-modules.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A human-vehicle-road-cloud intelligent network algorithm testing system, characterized in that: include: Road data module, real-time system module, driving simulation module and status acquisition module; The road data module collects the road condition data of the current road in real time, filters out the road scene data and the road vehicle data from the road condition data, and sends them to the real-time system module; The real-time system module tests the intelligent network-connected regulation and control algorithm to be tested based on the road scene data, the road vehicle data, the vehicle interface data of the virtual vehicle to be tested, and the vehicle dynamics model of the virtual vehicle to be tested, obtains the motion control information of the virtual vehicle to be tested and sends it to the driving simulation module, wherein the intelligent network-connected regulation and control algorithm to be tested is used to perform lateral, longitudinal and vertical motion control on the virtual vehicle to obtain the motion control information of the virtual vehicle; The driving simulation module performs vehicle dynamic simulation based on the motion control information; The state acquisition module collects driving experience information of the driver and passengers when the driving simulation module performs vehicle dynamic simulation, evaluates the intelligent network connection regulation and control algorithm to be tested based on the driving experience information to obtain a first evaluation result, evaluates the intelligent network connection regulation and control algorithm to be tested based on the motion control information to obtain a second evaluation result, and obtains a final evaluation result based on the first evaluation result and the second evaluation result.

2. The human-vehicle-road-cloud intelligent network algorithm testing system according to claim 1, characterized in that: The road condition data at least includes road scene data of the current road, road vehicle data, road construction information, road accident information and road congestion information.

3. The human-vehicle-road-cloud intelligent network algorithm testing system according to claim 1, characterized in that: The real-time system module includes a host computer and a real-time machine that are communicatively connected to each other, wherein the host computer is provided with test management software, automated test software and a vehicle dynamics model of the virtual vehicle to be tested; The test management software deploys the automated test software and the vehicle dynamics model of the virtual vehicle to be tested to the real-time machine before testing; The real-time machine tests the intelligent connected regulation and control algorithm to be tested based on the road scene data, the road vehicle data, the whole vehicle interface data of the virtual vehicle to be tested, and the vehicle dynamics model of the virtual vehicle to be tested through the automated testing software.

4. The human-vehicle-road-cloud intelligent network algorithm testing system according to claim 3, characterized in that: The real-time system module also includes a communication module that is communicatively connected to the real-time machine; The real-time machine sends the motion control information of the virtual vehicle to be tested to the driving simulation module through the communication module.

5. The human-vehicle-road-cloud intelligent network algorithm testing system according to claim 4, characterized in that: The real-time system module also includes a data simulation module that is communicatively connected to the communication module; The data simulation module configures the vehicle interface data of the virtual vehicle to be tested and sends it to the real-time machine through the communication module.

6. The human-vehicle-road-cloud intelligent network algorithm testing system according to claim 1, characterized in that: The motion control information includes lateral motion control information, longitudinal motion control information and vertical motion control information of the vehicle.

7. The human-vehicle-road-cloud intelligent network algorithm testing system according to claim 1, characterized in that: The state acquisition module includes a camera, a wearable device and a data analysis unit; The camera collects the facial expression information and eye state information of the driver and passenger during the vehicle dynamic simulation by the driving simulation module and sends the information to the data analysis unit; The wearable device collects blood oxygen concentration information of a target area of ​​the cerebral cortex of a driver or passenger during a vehicle dynamic simulation by the driving simulation module and sends the information to the data analysis unit; The data analysis unit evaluates the intelligent network-connected regulation and control algorithm to be tested based on the facial expression information, the eye state information and the blood oxygen concentration information to obtain a first evaluation result.

8. The human-vehicle-road-cloud intelligent network algorithm testing system according to claim 7, characterized in that: The data analysis unit determines whether the driver is in one of the following states: fatigue, tension, panic and excitement based on the facial expression information, the eye state information and the blood oxygen concentration information. If not, the first evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent network-connected regulation and control algorithm to be tested is reasonable. If so, the first evaluation result is determined to be that the motion control of the virtual vehicle by the intelligent network-connected regulation and control algorithm to be tested is unreasonable.

9. The human-vehicle-road-cloud intelligent network algorithm testing system according to claim 8, characterized in that: The data analysis unit determines whether the motion control information is motion control information for normal driving. If so, it determines that the second evaluation result is that the motion control of the virtual vehicle by the intelligent networked regulation and control algorithm to be tested is reasonable. If not, it determines that the second evaluation result is that the motion control of the virtual vehicle by the intelligent networked regulation and control algorithm to be tested is unreasonable.

10. The human-vehicle-road-cloud intelligent network algorithm testing system according to claim 9, characterized in that: When both the first evaluation result and the second evaluation result indicate that the motion control of the virtual vehicle by the intelligent connected control algorithm to be tested is reasonable, the data analysis unit determines that the final evaluation result is passed.

Images