In-vehicle safety officer takeover ability training system and method
Through the in-vehicle safety officer takeover capability training system that simulates actual driving scenarios, the takeover behavior is collected and evaluated in real time, the skill transfer problem in the existing training methods is solved, the safety officer takeover capability is improved, and the safety of autonomous driving vehicles is ensured.
Patent Information
- Application Number
- CN202510850846.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-24
- Publication Date
- 2025-08-08
AI Technical Summary
The existing training methods for taking over safety officers in-vehicles cannot fully meet the needs of cultivating high-quality autonomous driving in-vehicle safety officers, especially in actual scenarios, it is difficult to achieve effective transfer of skills and timely and accurate takeover operations.
It provides a vehicle safety officer takeover capability training system, including an autonomous driving takeover capability training platform, a site, a simulation unit, an RFID tag information reading unit, a takeover instruction control unit and a capability evaluation unit. By simulating the actual driving scenario, the safety officer's takeover behavior is collected and evaluated in real time and the takeover capability score is calculated.
The safety officer's takeover ability in different autonomous driving scenarios has been systematically improved, ensuring the safe operation of autonomous driving vehicles, and achieving standardized and skilled takeover operations.
Smart Images

Figure CN120452285A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of autonomous driving capability training, and in particular relates to a system and method for training in-vehicle safety officer takeover capability. Background Art
[0002] Currently, the development of autonomous driving technology is still in its infancy. Limited by factors such as technological maturity, road complexity, and regulatory maturity, autonomous vehicles must be equipped with a safety officer during testing or commercial operation to monitor the vehicle's operating status in real time. As the last line of defense for ensuring the safety of autonomous vehicles, in-vehicle safety officers not only require solid theoretical knowledge and operational expertise, but also rapid response and precise takeover skills.
[0003] Currently, mainstream training methods for in-car safety officers to take over include video tutorials, classroom instruction, and simulated driving. While theoretical training methods like video tutorials or classroom instruction can impart basic knowledge and operational specifications, the lack of immersive training in real-world scenarios often makes it difficult for trainees to effectively transfer these skills to actual operations. Simulated driving training, while providing practical opportunities, can also lead to delays and errors in taking over due to nervousness, as the simulated environment differs from real-world road conditions.
[0004] It can be seen that the existing training methods cannot fully meet the needs of cultivating high-quality in-vehicle safety officers for autonomous driving. Therefore, it is urgent to propose a system and method for training in-vehicle safety officers' takeover capabilities. Summary of the Invention
[0005] In order to solve the above technical problems, the present invention proposes a system and method for training the in-vehicle safety officer's takeover ability. Through systematic training methods and professional training equipment, the proficiency and standardization of the in-vehicle safety officer's takeover operation are improved, ensuring the safe operation of autonomous driving vehicles.
[0006] To achieve the above objectives, the present invention provides an in-vehicle safety officer takeover capability training system, comprising: an autonomous driving takeover capability training platform, an autonomous driving takeover capability training venue, a takeover training scenario simulation unit, an RFID tag information reading unit, a takeover training scenario control unit, a takeover instruction control unit, a takeover behavior collection unit, and a takeover capability evaluation unit;
[0007] The autonomous driving takeover capability training platform is used to provide a training environment and operation objects;
[0008] The autonomous driving takeover capability training venue is used to set training scenarios and trigger conditions for the in-vehicle safety officer takeover capability training;
[0009] The takeover training scenario simulation unit is used to cooperate with the autonomous driving takeover capability training venue to set up a designed autonomous driving takeover training scenario;
[0010] The RFID tag information reading unit is connected to the takeover training scenario control unit and the takeover instruction control unit, and is used to read the information in the RFID tag;
[0011] The takeover training scene control unit is used to control the takeover training scene simulation props to start running according to the trigger condition;
[0012] The takeover instruction control unit is used to issue a takeover request signal to the safety officer in the vehicle according to the trigger condition;
[0013] The takeover behavior collection unit is connected to the takeover instruction control unit and is used to collect the takeover behavior data of the security officer in the vehicle in real time after receiving the takeover request release signal;
[0014] The takeover capability evaluation unit is connected to the takeover behavior collection unit, and is used to receive and store the takeover behavior data of the in-vehicle safety officer sent by the takeover behavior collection unit, and calculate and output the takeover capability evaluation score of the in-vehicle safety officer.
[0015] Optionally, the autonomous driving takeover capability training platform is an autonomous driving vehicle with anti-rollover devices installed on the two rear wheels of the vehicle, and a voice transmission system and a co-pilot control system installed on the vehicle. The instructor can remotely provide guidance to the safety officer in the vehicle through voice at any time or take over control of the vehicle through the co-pilot control system.
[0016] Optionally, a remote monitoring unit is also included, which is connected to the automatic driving takeover capability training platform via a wireless network, and is used for the instructor to remotely monitor the operating behavior of the safety officer in the car in real time. When the safety officer in the car does not know how to take over, the instructor remotely guides the safety officer in the car to operate in real time through the voice transmission system; when the automatic driving takeover capability training platform has the possibility of a collision, the instructor remotely takes over and controls the operation status of the automatic driving takeover capability training platform through the co-pilot control system.
[0017] Optionally, the autonomous driving takeover capability training site is a section of intelligent driving test road within a closed training area. The road is divided into multiple training areas based on autonomous driving takeover capability training scenarios. A training area identification device is fixed at the beginning of each training area, and an RFID tag is installed in the training area identification device. RFID tags are also embedded in the road surface at designated locations in each training area. The RFID tags are inscribed with the takeover training scenario number and the code for issuing a takeover request for each training area.
[0018] Optionally, the RFID tag information reading unit is an RFID reader / writer, which is connected to the takeover training scenario control unit and the takeover instruction control unit, and is used to read the takeover training scenario number information and the code information requiring a takeover request in the RFID tag, and send the information to the takeover training scenario control unit or the takeover instruction control unit.
[0019] Optionally, the takeover training scenario control unit is used to start the takeover training scenario simulation props deployed in the autonomous driving takeover capability training field to automatically run along the set path according to the received takeover training scenario number information.
[0020] Optionally, the takeover behavior collection unit includes a sensor, which collects the takeover behavior data of the hands or feet of the safety officer in the vehicle in real time after receiving the takeover request release signal.
[0021] Optionally, the takeover capability evaluation unit includes a data processing subunit and an evaluation subunit:
[0022] The data processing subunit is used to receive and store the takeover behavior data sent by the takeover behavior collection unit, and extract key behavior features from the takeover behavior data, wherein the key behavior features include takeover speed, takeover performance, etc.;
[0023] The evaluation subunit is used to analyze and process the extracted key behavioral features, and calculate and output an evaluation score of the takeover capability of the in-vehicle safety officer.
[0024] The present invention also provides a method for training the in-vehicle safety officer's ability to take over, comprising the following steps:
[0025] The instructor selects the required autonomous driving takeover training scenario from the autonomous driving takeover training scenario library based on the training progress of the in-vehicle safety officer, and sets up the takeover training scenario simulation props at the autonomous driving takeover capability training venue according to the needs of the takeover training scenario;
[0026] Based on the selected autonomous driving takeover training scenario, the instructor writes the takeover training scenario number that matches the takeover training scenario in the training area on the RFID tag in the training area identification device at the beginning of the corresponding training area of the autonomous driving takeover capability training venue, and writes the code for issuing a takeover request on the RFID tag at the designated location of the corresponding training area;
[0027] The instructor explains in detail to the safety officer the typical takeover scenarios of autonomous vehicles, the form of takeover requests, the standard takeover methods, and vehicle feedback.
[0028] After the in-car safety officer has learned about typical takeover scenarios, takeover request formats, standardized takeover methods, and vehicle feedback for autonomous vehicles, the autonomous driving takeover capability training platform is activated. The training platform automatically drives around the autonomous driving takeover capability training site according to the destination and route set by the in-car safety officer.
[0029] When the RFID tag information reading unit reads the takeover training scene number information in the RFID tag in the training area identification device fixed on the road surface at the starting end of each training area, the number information is sent to the takeover training scene control unit;
[0030] The takeover training scenario control unit matches the number information with the takeover training scenario, and according to the takeover training scenario content, starts the takeover training scenario simulation props deployed in the area to run along the set path;
[0031] When the RFID tag information reading unit reads the code information of the RFID tag buried at the designated position of the training area and requires issuing a takeover request, the code information is sent to the takeover instruction control unit;
[0032] When the takeover command control unit receives the code information for issuing a takeover request, it issues a takeover request to the safety officer in the vehicle;
[0033] After hearing the takeover request, the safety officer in the vehicle takes over the vehicle through the switch button, brake pedal, steering wheel or lever operating parts;
[0034] When the in-vehicle security officer performs the takeover operation, the takeover behavior collection unit collects the takeover behavior data of the in-vehicle security officer's hands or feet in real time, and sends the collected takeover behavior data to the takeover capability evaluation unit for storage;
[0035] After the autonomous driving takeover capability training platform reaches the destination according to the driving destination and route set by the in-vehicle safety officer, the training is completed. The takeover capability evaluation unit analyzes and processes the received takeover behavior data and calculates the in-vehicle safety officer's takeover capability evaluation score.
[0036] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method.
[0037] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when the computer program is executed by a processor.
[0038] Compared with the prior art, the present invention has the following advantages and technical effects:
[0039] The in-vehicle safety officer takeover capability training system and method provided by the present invention can systematically improve the in-vehicle safety officer's takeover capability in different autonomous driving scenarios during the operation of the autonomous driving vehicle, so that the officer can perform takeover operations in a standardized, skilled, timely and accurate manner, effectively ensuring the safe operation of the autonomous driving vehicle, which is of great significance for promoting the application of autonomous driving technology. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:
[0041] Figure 1 This is a structural diagram of an in-vehicle safety officer takeover capability training system according to an embodiment of the present invention, including: 1. remote monitoring unit; 2. autonomous driving takeover capability training platform; 3. autonomous driving takeover capability training venue; 4. takeover training scenario simulation unit; 5. RFID tag information reading unit; 6. takeover training scenario control unit; 7. takeover instruction control unit; 8. takeover behavior collection unit; 9. takeover capability evaluation unit;
[0042] Figure 2 This is a training process for in-vehicle safety officer takeover capabilities according to an embodiment of the present invention. DETAILED DESCRIPTION
[0043] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0044] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0045] Example 1
[0046] This embodiment proposes a vehicle safety officer takeover capability training system, such as Figure 1 As shown, it specifically includes: a remote monitoring unit 1, an autonomous driving takeover capability training platform 2, an autonomous driving takeover capability training site 3, a takeover training scenario simulation unit 4, an RFID tag information reading unit 5, a takeover training scenario control unit 6, a takeover instruction control unit 7, a takeover behavior collection unit 8, and a takeover capability evaluation unit 9.
[0047] Remote monitoring unit 1 is connected to autonomous driving takeover capability training platform 2 via a wireless network, allowing the instructor to remotely monitor the safety officer's operating behavior in real time. If the safety officer is unsure how to take over, the instructor provides real-time guidance via a voice transmission system. If the autonomous driving takeover capability training platform indicates a potential collision, the instructor remotely takes over control of the platform's operation via the co-pilot control system.
[0048] The autonomous driving takeover capability training platform 2 is an autonomous driving vehicle with anti-rollover devices installed on its two rear wheels. The vehicle is also equipped with a voice transmission system and a co-pilot control system. The instructor can remotely provide guidance to the safety officer in the vehicle through voice at any time or take over control of the vehicle through the co-pilot control system.
[0049] The autonomous driving takeover capability training venue 3 is used to train the takeover capability of the in-vehicle safety officer. The autonomous driving takeover capability training platform 2 can drive autonomously at the autonomous driving takeover capability training venue 3.
[0050] Furthermore, the autonomous driving takeover capability training site 3 is a section of intelligent driving test road within a closed training site. Typical highway traffic scenes and urban road traffic scenes are set up on the road, and vehicle-road communication infrastructure, traffic signal facilities, various meteorological and tunnel simulation equipment, etc. are installed.
[0051] Furthermore, based on the traffic scenario set up at the autonomous driving takeover capability training site 3 and the designed autonomous driving takeover training scenario, the training site 3 is divided into multiple training areas. A training area identification device is fixed at the beginning of each training area, and an RFID tag is installed in the training area identification device. At the same time, an RFID tag is also buried on the road surface at a designated location in each training area.
[0052] Specifically, a designated location in the training area where a takeover request is required is determined based on the designed autonomous driving takeover training scenario.
[0053] An RFID reader / writer writes the number corresponding to the autonomous driving takeover training scenario for that training area onto the RFID tag within the training area identification device. The code for issuing the takeover request is also written onto the RFID tag embedded in a designated location within each training area. The number and code on the RFID tag can be dynamically adjusted via the RFID reader / writer. Code encoding rules can include, but are not limited to, JG, 10, etc.
[0054] The takeover training scenario simulation unit 4 is used to coordinate with the autonomous driving takeover capability training site 3 to create the designed autonomous driving takeover training scenario. This unit includes, but is not limited to, simulation props for the takeover training scenario, such as dummies, dummy vehicles, cones, and high-voltage pylons. The dummies and dummy vehicles can be wirelessly connected to the takeover training scenario control unit 6 and operate according to the path configured by the takeover training scenario control unit 6.
[0055] The RFID tag information reading unit 5 is an RFID reader / writer connected to the takeover training scenario control unit 6 and the takeover instruction control unit 7. When the autonomous driving device operation capability training platform 2 passes by the training area identification device or a designated location in the training area, it automatically reads the takeover training scenario number information from the RFID tag in the training area identification device or the code information of the takeover request from the RFID tag embedded in the designated location of each training area.
[0056] Furthermore, if the RFID tag information reading unit 5 reads the takeover training scenario number information, it sends the information to the takeover training scenario control unit 6; if it reads the code information requiring a takeover request, it sends the information to the takeover instruction control unit 7.
[0057] The takeover training scene control unit 6 is used to receive the takeover training scene number information sent by the RFID tag information reading unit 5, and then match the takeover training scene number information with the autonomous driving takeover training scene, and start the takeover training scene simulation props deployed on the autonomous driving takeover capability training site 3 to automatically run according to the set path.
[0058] The takeover instruction control unit 7 is used to receive the code information sent by the RFID tag information reading unit 5 requesting a takeover request, and then send a takeover request to the safety officer in the vehicle in the form of voice or vibration according to the code information, and send the information of issuing the takeover request to the takeover behavior collection unit 8.
[0059] The takeover behavior collection unit 8 is connected to the takeover instruction control unit 7 and is used to collect the takeover behavior data of the security officer in the vehicle in real time after receiving the takeover request signal from the takeover instruction control unit;
[0060] The takeover capability evaluation unit 9 is connected to the takeover behavior collection unit 8 and is configured to receive and store the takeover behavior data of the onboard safety officer sent by the takeover behavior collection unit 8, storing it in real time in a takeover behavior database. After the training is completed, the takeover capability evaluation unit 9 reads the takeover behavior data from the takeover behavior database, analyzes and processes it, and calculates and outputs a takeover capability evaluation score for the onboard safety officer.
[0061] Specifically, the RFID tag information reading unit 5, the takeover training scenario control unit 6, the takeover instruction control unit 7, the takeover behavior collection unit 8 and the takeover capability evaluation unit 9 are all installed on the automatic driving device operation capability training platform 2.
[0062] As an additional implementation method, the specific process of designing the autonomous driving takeover training scenario in this embodiment is as follows:
[0063] (1) Record the traffic environment, road conditions, vehicle operating conditions, and other information each time the autonomous vehicle issues a takeover request during actual operation. Summarize and describe this information to form an autonomous driving takeover scenario. For example, an autonomous driving takeover scenario can be described as follows: On a sunny day with low traffic flow, a vehicle is traveling at a speed of 50 km / h to a flat intersection. When it reaches the middle of the intersection, the vehicle issues a takeover request due to the missing lane line.
[0064] (2) After the recorded autonomous driving takeover scenarios are stored in the database, an autonomous driving takeover scenario library is constructed.
[0065] (3) In the Python integrated development environment, use the Python programming language to write a program that automatically reads the scenes in the autonomous driving takeover scene library and calls the pre-trained deep learning models in AI models such as Volcano Engine and Doubao, so that the deep learning model can automatically understand all the scenes in the autonomous driving takeover scene library and extract typical takeover scenes.
[0066] (4) After the deep learning model extracts typical takeover scenarios, each typical takeover scenario is modified according to the training requirements to design corresponding autonomous driving takeover training scenarios;
[0067] (5) All designed autonomous driving takeover training scenarios are numbered. The numbering rules may be, but are not limited to, S1-1, S1-2, S2-1, etc., and then these scenarios are stored in a database to construct an autonomous driving takeover training scenario library.
[0068] Example 2
[0069] This embodiment also provides a method for training an in-vehicle safety officer's ability to take over, the method comprising:
[0070] (1) Takeover training scenario layout. The instructor selects the required autonomous driving takeover training scenario from the autonomous driving takeover training scenario library based on the training progress of the in-vehicle safety officer, and arranges takeover training scenario simulation props such as dummies and dummy cars on the autonomous driving takeover capability training site according to the needs of the takeover training scenario.
[0071] (2) Writing the takeover training scenario number and the code for issuing a takeover request. Based on the selected autonomous driving takeover training scenario, the instructor writes the takeover training scenario number that matches the takeover training scenario of the training area on the RFID tag in the training area identification device at the starting end of the corresponding training area of the autonomous driving takeover capability training venue, and writes the code for issuing a takeover request on the RFID tag buried in the designated location of each training area.
[0072] (3) Explanation of autonomous driving takeover knowledge and standard operations. The instructor explains in detail to the safety officer in the car the typical takeover scenarios of autonomous driving vehicles, the form of takeover requests, standard takeover methods, vehicle feedback, etc.
[0073] (4) The autonomous driving takeover capability training platform starts driving. After the in-vehicle safety officer starts the autonomous driving takeover capability training platform, the training platform automatically drives around the autonomous driving takeover capability training site according to the set driving destination and route.
[0074] (5) RFID tag information reading: When driving to each training area in the training site, the RFID tag information reading unit reads the takeover training scene number information and the code information of the takeover request in the training area.
[0075] (6) Startup of the takeover training scenario simulation props. When the takeover training scenario control unit receives the takeover training scenario number information sent by the RFID tag information reading unit, it starts the takeover training scenario simulation props deployed on the autonomous driving takeover capability training site to automatically run along the set path.
[0076] (7) Takeover request triggering: When the takeover instruction control unit receives the code information of the takeover request sent by the RFID tag information reading unit, it triggers the takeover request.
[0077] (8) Implementation of takeover. After hearing the takeover request, the in-vehicle safety officer implements the takeover operation through operating components such as switch buttons, brake pedals, steering wheels, and levers.
[0078] (9) Takeover behavior data collection. When the in-vehicle safety officer performs the takeover operation, the takeover behavior collection unit collects the takeover behavior data of the in-vehicle safety officer's hands or feet in real time and sends the collected takeover behavior data to the takeover capability evaluation unit.
[0079] (10) Storage of takeover behavior data: The takeover capability evaluation unit receives the information sent by the takeover behavior data collection unit in real time and stores it.
[0080] (11) Takeover capability evaluation. Repeat steps (4) to (10). After the autonomous driving takeover capability training platform reaches the destination and route set by the in-vehicle safety officer, the takeover capability evaluation unit analyzes and processes the received takeover behavior data and calculates the in-vehicle safety officer's takeover capability evaluation score.
[0081] As a specific implementation method, a method for training the in-vehicle safety officer's ability to take over automatic driving is as follows: Figure 2 As shown, the specific process includes the following:
[0082] P1: The instructor selects the required autonomous driving takeover training scenario from the autonomous driving takeover training scenario library based on the training progress of the in-vehicle safety officer, and deploys dummies, dummy cars and other takeover training scenario simulation props on the autonomous driving takeover capability training site according to the needs of the takeover training scenario.
[0083] P2: Based on the selected autonomous driving takeover training scenario, the instructor writes the takeover training scenario number matching the training area onto the RFID tag embedded in the training area identification device at the beginning of the corresponding training area in the autonomous driving takeover capability training venue. The instructor also writes the code for issuing a takeover request onto the RFID tag embedded in a designated location in each training area.
[0084] P3: Based on the stationary autonomous driving takeover capability training platform 2, the instructor explains in detail to the safety officer in the vehicle the typical takeover scenarios of autonomous driving vehicles, the form of takeover requests, the standardized takeover methods, vehicle feedback, and other knowledge.
[0085] P4: After the in-car safety officer has learned about typical takeover scenarios, takeover request formats, standardized takeover methods, and vehicle feedback, the autonomous driving takeover capability training platform 2 is launched. After the destination and route are set and the autonomous driving function is activated, the autonomous driving takeover capability training platform 2 begins automatically driving around the autonomous driving takeover capability training site 3 according to the destination and route set by the in-car safety officer.
[0086] P5: When the autonomous driving takeover capability training platform 2 is automatically driving in the autonomous driving takeover capability training site 3, when the RFID tag information reading unit 5 reads the takeover training scene number information in the RFID tag in the training area identification device fixed on the road surface at the starting end of each training area, the number information is sent to the takeover training scene control unit 6.
[0087] P6: The takeover training scenario control unit 6 matches the number information with the takeover training scenario, and then starts the takeover training scenario simulation props deployed in the area to run along the set path according to the takeover training scenario content.
[0088] P7: The autonomous driving takeover capability training platform 2 continues to travel in the training area. When the RFID tag information reading unit 5 reads the code information of the takeover request in the RFID tag buried at the designated location of the training area, it sends the code information to the takeover instruction control unit 7.
[0089] P8: When the takeover instruction control unit 7 receives the code information requesting a takeover request, it sends a takeover request to the safety officer in the vehicle by voice, vibration, etc.
[0090] P9: After hearing the takeover request, the safety officer in the car takes over the vehicle through operating components such as switch buttons, brake pedals, steering wheel, and levers.
[0091] P10: When the in-vehicle safety officer performs the takeover operation, the takeover behavior collection unit collects the takeover behavior data of the in-vehicle safety officer's hands or feet in real time, and sends the collected takeover behavior data to the takeover capability evaluation unit for storage.
[0092] P11: Repeat steps P5-P10. After the autonomous driving takeover capability training platform reaches the destination and route set by the in-vehicle safety officer, the training is completed. The takeover capability evaluation unit analyzes and processes the received takeover behavior data and calculates the in-vehicle safety officer's takeover capability evaluation score.
[0093] Furthermore, the takeover behavior data in the takeover operation includes real-time collection of the moment when the safety officer in the vehicle operates the switch buttons, brake pedal, steering wheel, lever and other operating components after the takeover instruction control unit issues a takeover request, as well as the operating status of the vehicle after the operation, to form a takeover behavior database for subsequent evaluation and analysis.
[0094] Furthermore, when the in-car safety officer is performing the above operations, the instructor can check the in-car safety officer's takeover status in real time through the remote monitoring unit. When the in-car safety officer does not know how to take over, the instructor can remotely guide the in-car safety officer to perform the operation in real time through the voice transmission system; when the automatic driving takeover capability training platform has the possibility of a collision, the instructor can remotely take over and control the operation status of the automatic driving takeover capability training platform through the co-pilot control system.
[0095] Furthermore, the takeover capability evaluation unit extracts multiple key behavioral features from the in-vehicle safety officer takeover behavior database, including but not limited to takeover speed, takeover performance, etc. Furthermore, based on the in-vehicle safety officer takeover behavior database, the specific process for calculating takeover speed and takeover performance is as follows:
[0096] (1) The takeover behavior dataset D in the in-vehicle safety officer takeover behavior database is:
[0097] D={D1,D2,…,D i ,…,D m},
[0098] Among them, D i is the takeover behavior dataset of the i-th in-vehicle safety officer; m is the total number of in-vehicle safety officers who participated in the training.
[0099] (2) Dataset D of the takeover behavior of the i-th in-car safety officer i for:
[0100] D i ={D i1 ,D i2 ,…,D ik ,…,D ip},
[0101] Among them, D ik is the takeover behavior dataset of the i-th in-vehicle safety officer during the k-th training; p is the total number of training times.
[0102] (3) The takeover behavior dataset D of the i-th in-vehicle safety officer during the k-th training ik for:
[0103]
[0104] Among them: ts ikj is the time when the jth takeover instruction is issued by the i-th in-vehicle safety officer during the k-th training; te ikj is the time when the i-th in-car safety officer starts to take over after the j-th takeover instruction is issued during the k-th training; p ikj is the operating status of the vehicle after the i-th in-vehicle safety officer takes over for the j-th time during the k-th training, where 0 indicates normal operation and 1 indicates a collision. n is the total number of times the i-th in-vehicle safety officer takes over during the k-th training.
[0105] (4) Calculate the takeover speed T of the i-th in-vehicle safety officer during the k-th training ik :
[0106]
[0107] Where: T ik is the takeover speed of the i-th in-vehicle safety officer during the k-th training; I(p ikj =1) is an indicator function, when the condition p ikj When =1 holds, the function value is 1, otherwise it is 0.
[0108] (5) Calculate the takeover performance P of the i-th in-vehicle safety officer during the k-th training ik :
[0109]
[0110] Where: P ik is the takeover performance of the i-th in-vehicle safety officer during the k-th training.
[0111] Furthermore, after the key behavioral features are extracted, they are input into the preset training evaluation model for calculation to obtain the takeover capability evaluation score. The takeover capability evaluation score is a quantitative indicator, including but not limited to being expressed on a scale of 0 to 100, and can also be converted into a scale of 0 to 5 or 0 to 10, etc., reflecting the overall takeover level of the in-vehicle safety officer. The specific process of calculating the takeover capability evaluation score of the k-th training of the i-th in-vehicle safety officer using the preset takeover capability evaluation model is as follows:
[0112] (1) Normalize the takeover speed of the i-th in-vehicle safety officer during the k-th training:
[0113]
[0114] Where: T' ik is the normalized takeover speed of the i-th in-vehicle safety officer during the k-th training; T min The maximum value of the takeover speed is set; T min The minimum takeover speed.
[0115] (2) Calculate the takeover ability evaluation score S of the kth training of the i-th in-vehicle safety officer ik :
[0116] S ik =(w A *P ik +w B *T′ ik )*100,
[0117] Where: S ik w is the takeover ability evaluation score of the i-th in-vehicle safety officer in the k-th training; A is the takeover performance weight set; w B The takeover speed weight is set.
[0118] Example 3
[0119] This embodiment further discloses a computer device, including a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the steps of the method of embodiment 1.
[0120] Example 4
[0121] This embodiment further discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method of embodiment 1 are implemented.
[0122] Example 5
[0123] This embodiment further discloses a computer program product, including a computer program, which implements the steps of the method of embodiment 1 when executed by a processor.
[0124] The above are merely preferred embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.
Claims
1. A vehicle safety officer takeover capability training system, characterized in that: include: Autonomous driving takeover capability training platform, autonomous driving takeover capability training venue, takeover training scenario simulation unit, RFID tag information reading unit, takeover training scenario control unit, takeover instruction control unit, takeover behavior collection unit, and takeover capability evaluation unit; The autonomous driving takeover capability training platform is used to provide a training environment and operation objects; The autonomous driving takeover capability training venue is used to set training scenarios and trigger conditions for the in-vehicle safety officer takeover capability training; The takeover training scenario simulation unit is used to cooperate with the autonomous driving takeover capability training venue to set up a designed autonomous driving takeover training scenario; The RFID tag information reading unit is connected to the takeover training scenario control unit and the takeover instruction control unit, and is used to read the information in the RFID tag; The takeover training scene control unit is used to control the takeover training scene simulation props to start running according to the trigger condition; The takeover instruction control unit is used to issue a takeover request to the in-vehicle safety officer according to the trigger condition; The takeover behavior collection unit is connected to the takeover instruction control unit and is used to collect the takeover behavior data of the security officer in the vehicle in real time after receiving the takeover request release signal; The takeover capability evaluation unit is connected to the takeover behavior collection unit, and is used to receive and store the takeover behavior data of the in-vehicle safety officer sent by the takeover behavior collection unit, and calculate and output the takeover capability evaluation score of the in-vehicle safety officer.
2. The in-vehicle safety officer takeover capability training system according to claim 1, characterized in that: The autonomous driving takeover capability training platform is an autonomous driving vehicle with anti-rollover devices installed on its two rear wheels, and a voice transmission system and a co-pilot control system. The instructor can remotely provide guidance to the safety officer in the vehicle through voice at any time or take over control of the vehicle through the co-pilot control system.
3. The in-vehicle safety officer takeover capability training system according to claim 1, characterized in that: The remote monitoring unit is connected to the autonomous driving takeover capability training platform via a wireless network, and is used by the instructor to remotely monitor the operating behavior of the in-car safety officer in real time. If the in-car safety officer does not know how to take over, the instructor can remotely guide the in-car safety officer in real time through the voice transmission system; When there is a possibility of collision in the autonomous driving takeover capability training platform, the instructor remotely takes over the operation of the autonomous driving takeover capability training platform through the co-pilot control system.
4. The in-vehicle safety officer takeover capability training system according to claim 1, characterized in that: The autonomous driving takeover capability training site is a section of intelligent driving test road within a closed training site, and the road is divided into multiple training areas according to the training scenarios of autonomous driving takeover capability. A training area identification device is fixed at the beginning of each training area, and an RFID tag is installed in the training area identification device. At the same time, an RFID tag is also buried on the road surface at a designated location in each training area. The RFID tag is written with the takeover training scenario number information of each training area and the code information requiring a takeover request.
5. The in-vehicle safety officer takeover capability training system according to claim 1, characterized in that: The RFID tag information reading unit is an RFID reader / writer, which is used to read the takeover training scenario number information and the code information requiring the takeover request in the RFID tag, and send the information to the takeover training scenario control unit or the takeover instruction control unit.
6. The in-vehicle safety officer takeover capability training system according to claim 1, characterized in that: The takeover training scenario control unit is used to start the takeover training scenario simulation props arranged on the automatic driving takeover capability training site to automatically run along the set path according to the received takeover training scenario number information.
7. The in-vehicle safety officer takeover capability training system according to claim 1, characterized in that: The takeover behavior collection unit includes a sensor, which collects the takeover behavior data of the hands or feet of the safety officer in the vehicle in real time after receiving the takeover request release signal.
8. The in-vehicle safety officer takeover capability training system according to claim 1, characterized in that: The takeover capability evaluation unit includes a data processing subunit and an evaluation subunit: The data processing subunit is used to receive and store the takeover behavior data sent by the takeover behavior collection unit, and extract key behavior features from the takeover behavior data, wherein the key behavior features include takeover speed and takeover performance; The evaluation subunit is used to analyze and process the extracted key behavioral features, and calculate and output an evaluation score of the takeover capability of the in-vehicle safety officer.
9. A method for training the ability of in-vehicle safety officers to take over, characterized in that: The system according to any one of claims 1 to 8 comprises the following steps: The instructor selects the required autonomous driving takeover training scenario from the autonomous driving takeover training scenario library based on the training progress of the in-vehicle safety officer, and sets up the takeover training scenario simulation props at the autonomous driving takeover capability training venue according to the needs of the takeover training scenario; Based on the selected autonomous driving takeover training scenario, the instructor writes the takeover training scenario number that matches the takeover training scenario in the training area on the RFID tag in the training area identification device at the beginning of the corresponding training area of the autonomous driving takeover capability training venue, and writes the code for issuing a takeover request on the RFID tag at the designated location of the corresponding training area; The instructor explains in detail to the safety officer the typical takeover scenarios of autonomous vehicles, the form of takeover requests, the standard takeover methods, and vehicle feedback. After the in-car safety officer understands the typical takeover scenarios, takeover request formats, standardized takeover methods, and vehicle feedback for autonomous vehicles, the autonomous driving takeover capability training platform is activated. The training platform automatically drives around the autonomous driving takeover capability training site according to the destination and route set by the in-car safety officer. When the RFID tag information reading unit reads the takeover training scene number information in the RFID tag in the training area identification device fixed on the road surface at the starting end of each training area, the number information is sent to the takeover training scene control unit; The takeover training scenario control unit matches the number information with the takeover training scenario, and according to the takeover training scenario content, starts the takeover training scenario simulation props deployed in the training area to run along the set path; When the RFID tag information reading unit reads the code information of the RFID tag buried at the designated position of the training area and requires issuing a takeover request, the code information is sent to the takeover instruction control unit; When the takeover command control unit receives the code information for issuing a takeover request, it issues a takeover request to the safety officer in the vehicle; After hearing the takeover request, the safety officer in the vehicle takes over the vehicle through the switch button, brake pedal, steering wheel or lever operating parts; When the in-vehicle security officer performs the takeover operation, the takeover behavior collection unit collects the takeover behavior data of the in-vehicle security officer's hands or feet in real time, and sends the collected takeover behavior data to the takeover capability evaluation unit for storage; After the autonomous driving takeover capability training platform reaches the destination according to the driving destination and route set by the in-vehicle safety officer, the training is completed. The takeover capability evaluation unit analyzes and processes the received takeover behavior data and calculates the in-vehicle safety officer's takeover capability evaluation score.
10. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the method according to claim 9.