709 results about "Driving test" patented technology

The invention discloses an AR-fused remote driving vehicle virtual-real interaction hardware-in-the-loop test system, particularly relates to the technical field of automatic driving test, and is used for solving the problems of inaccurate coupling between a virtual scene and a real vehicle behavior and lack of AR prompt response evaluation. The method comprises the following steps: firstly, constructing a dynamic obstacle intention-driven prediction model based on time series data of a multi-modal sensor, and generating a trajectory probability distribution and risk thermodynamic diagram; then, space-time alignment of the virtual accident scene and the real environment is achieved through a dynamic binding algorithm, and the virtual-real shielding priority of an AR interface is dynamically adjusted; by simulating abnormal disturbance of a vehicle actuator, synchronously collecting control and watching responses of a driver, and extracting obstacle avoidance path deviation degree and takeover timeliness parameters; and finally, separating and compensating virtual and actual residual errors based on a path deviation index, realizing online correction of a virtual scene attitude and a dynamic trajectory, constructing a closed-loop optimization mechanism, and improving the precision and stability of a test system.

The invention relates to an automatic derivation method, system and device for an automatic driving test scene and a medium, and the method comprises the steps: analyzing a test demand of an automatic driving system, and mapping the test demand to a multi-dimensional scene element hierarchical architecture, so as to construct a multi-layer road scene; reasoning a scene condition dependency relationship by using a dual conditional probability, and generating an initial scene fragment set in combination with probability distribution sampling; establishing a multi-objective optimization function based on the scene type distribution weight, the element redundancy penalty factor, the physical consistency constraint and the security rule, screening and recombining the multiple initial scene fragment sets, and outputting an optimal scene set; extracting a causal logical relationship of the optimal scene set according to the directed acyclic graph, and generating a scene evolution path in combination with topological sorting; and executing multiple safety verification on the scene evolution path, and determining the optimal scene passing the verification as the automatic driving test scene. The technical problems that an existing test scene is low in generation efficiency, poor in authenticity and unreasonable in event chain logic are solved.

The invention belongs to the technical field of automatic driving testing, and particularly relates to an accident scene generation method based on a scene knowledge graph and considering accident causes. Comprising the following steps: step 1, modeling a scene knowledge graph; 2, modeling the scene graph time sequence prediction model; 3, modeling the time sequence causal inference model; step 4, modeling the scene graph time sequence decision generation model so as to generate an accident scene; according to the method, the accident scene database with high authenticity, high diversity and accident cause consistency can be constructed under the condition that the accident scene sample data size is limited, the number of accident scenes of the same type in the automatic driving algorithm closed-loop self-evolution cloud database is efficiently expanded, the constructed scene database is used for training the automatic driving algorithm, and the accuracy of the automatic driving algorithm is improved. The adaptive capacity of the self-driving automobile to scenes with the same type of accidents can be effectively enhanced, and safe and reliable operation of the self-driving automobile in the real world is guaranteed.

The invention discloses a safety device for reducing the back-up collision risk in the automatic driving test process, and belongs to the technical field of automatic driving test safety. Aiming at the collision risk problem caused by rapid movement of dynamic obstacles and communication delay when a vehicle backs up in a closed test site, the device obtains millimeter-level position coordinates and course angles through positioning units deployed on a test vehicle and a target object; a C-V2X protocol is utilized to construct a real-time data link to share vehicle position, speed, backing track prediction data and target object outer contour feature point coordinates, potential conflicts are identified based on future minimum spacing calculation and dynamic threshold, and a cooperative collision avoidance path is generated in combination with a distributed optimization algorithm. And the high-risk vehicle preferentially preempts a communication channel to issue deceleration, course adjustment and braking instructions to surrounding vehicles. The device can effectively improve the reversing safety in a multi-vehicle cooperative test scene, and is suitable for collision risk real-time early warning and active intervention of an automatic driving vehicle closed test site.

The invention belongs to the field of data processing, and discloses a test method for pedestrian trajectory prediction and emergency braking in an automatic driving platform. Comprising the steps of segmenting a real road environment in advance, obtaining grades of small road segments according to calculation of segmentation coefficients, carrying out resource allocation on the small segments according to the grades, and collecting multi-source heterogeneous data; importance optimization is introduced into a Transform model, possible prediction trajectories of individuals are generated, graph data are constructed by using the possible prediction trajectories, an Autoencoder anomaly detection mechanism is introduced, high-risk trajectories are screened out, and high-risk scenes are output; combining the road geometric model with the high-risk scene to form a scene library, loading the scene in a simulation platform, carrying out automatic driving test, carrying out virtual-real data cross validation in combination with real vehicle test data, and obtaining performance evaluation of an algorithm; pedestrian track prediction and emergency braking in the automatic driving process are achieved.

The invention relates to an online multi-dimensional comprehensive test evaluation method for an automatic driving test. The method comprises the following steps: constructing a test scene library; constructing a test scene complexity model; building an intelligent driving joint simulation test platform, and embedding the intelligent driving algorithm into the intelligent driving joint simulation test platform; constructing an automatic driving test multi-dimensional evaluation index system for performing multi-dimensional performance evaluation on the automatic driving vehicle; an intelligent driving joint simulation test platform is utilized to obtain vehicle test data, an evaluation index system and a test scene complexity model are combined to calculate objective evaluation index weights, then comprehensive performance quantitative scores are calculated, and finally an automatic evaluation report is output and obtained. Compared with the prior art, the subjectivity of manual evaluation can be reduced, and the evaluation efficiency of the automatic driving vehicle is improved; comprehensive and objective quantitative scoring can be carried out on the multi-dimensional performance of the automatic driving vehicle, and the influence of different test scenes on the scoring difference of the automatic driving vehicle is reduced.

The invention discloses an automatic driving test scene set optimization method, apparatus and device, and a storage medium. The method comprises the steps of generating a simulation scene file in an OpenSCENARIO format through preprocessing data; analyzing the simulation scene file through a teacher model, outputting a risk description text, receiving a scene feature vector and the risk description text through a student model, and outputting a failure probability; determining a comprehensive value index according to the failure probability, dynamically updating a scene library of automatic driving test scenes, collecting failure data in an AUT test, performing incremental fine tuning on the student model, and obtaining an optimized target test scene set; the problem of gradient estimation variance explosion caused by sparseness disasters can be effectively solved, the stability of model training is improved, the recognition capability of a rare long-tail failure scene is enhanced, the accuracy of failure probability prediction is improved, the judgment accuracy of the model to a boundary scene is improved, and the accuracy and comprehensiveness of scene value quantification are improved.

The invention discloses a multi-modal driving test scene generation method and system, and the method comprises the steps: obtaining a plurality of groups of model training samples, each group of model training samples comprising a user natural language description, a driving scene visual image, a risk point thermodynamic diagram, and a safety key scene description code; constructing a driving test scene generation framework, and training the driving test scene generation framework by using the multiple groups of model training samples to obtain a trained driving test scene generation framework; wherein the driving test scene generation framework comprises an image content coding model, a risk point mining model, a thermodynamic diagram coding model and a large language model, the risk point mining model and the thermodynamic diagram coding model are connected in series, and the output end of the image content coding model and the output end of the thermodynamic diagram coding model are connected with the input end of the large language model; and generating a multi-modal driving test scene according to the input user natural language description and the driving scene visual image by using the trained driving test scene generation architecture.

The invention belongs to the field of data analysis and processing, and discloses a virtuality and reality combined automobile automatic driving test method. Comprising the following steps: connecting a sensor suite and an automatic driving system of an automobile with a virtual twin center by using a CAN bus or an Ethernet interface, operating the automatic driving system, and generating a performance report; the method comprises the following steps: initializing raindrop attributes, setting the transparency of raindrops based on real environment data, introducing a complex wind field model and a surface ponding dynamic evolution mechanism, carrying out collision detection on all raindrop pairs by using a space segmentation algorithm, and simulating dynamic behaviors of the raindrops in the air based on a discrete element method and a smoothed particle fluid dynamics method. Visual rendering is carried out on the water spray effect; under the influence of raindrop physical characteristic modeling, output of laser radar point cloud scattering modeling, camera image fuzzy modeling, millimeter wave radar signal attenuation and scattering modeling and ultrasonic sensor interference modeling is iteratively updated, and automatic driving testing of an automobile under virtual-real combination is achieved.

The invention relates to the technical field of battery testing, in particular to an automatic BMS function testing system and method. The method comprises the following steps: acquiring BMS working parameters; performing short-circuit working condition configuration processing according to the BMS working parameters to generate short-circuit working condition configuration data; performing short-circuit signal simulation based on the short-circuit working condition configuration data to generate short-circuit simulation control signal time sequence data; based on the short-circuit analog control signal time sequence data, a battery pack current short-circuit driving test is carried out through a BMS monitoring system, and BMS monitoring response data is generated; and performing short-circuit current monitoring protection evaluation according to the BMS monitoring response data, and generating BMS function test level data. According to the invention, the automatic test of the BMS short-circuit protection function is realized by intelligently configuring the short-circuit working condition, and the efficiency, accuracy and repeatability of the test are remarkably improved.

The invention belongs to the field of automatic driving test, and provides an end-to-end automatic driving system closed-loop confrontation test method based on a world model, and the method comprises the steps: building an end-to-end automatic driving system test platform based on the world model, providing end-to-end automatic driving system perception input information, and carrying out the modeling of a camera sensor, the accurate alignment among the multi-modal scene information is realized; constructing a traffic flow module, and providing map and scene global state information; a traffic vehicle multi-agent collaborative confrontation strategy balancing strategy convergence and strategy diversity is designed, a dense reward function is constructed and matched with an agent heterogeneous strategy adaptive adjustment method, a global exploration target is planned as a whole, and the test efficiency is improved; all the modules are integrated, an end-to-end automatic driving system to be tested is integrated, and efficient closed-loop testing is achieved. According to the method, efficient testing of the performance of the end-to-end automatic driving system can be achieved, the robustness and safety of the end-to-end automatic driving system can be evaluated, and functional defects of the system can be rapidly excavated.

The invention discloses an automatic driving test case generation method based on a large model retrieval enhancement technology, and aims to improve the generation efficiency of an automatic driving test scene. The method comprises the following steps: firstly, cutting a preprocessed data text into blocks, and vectorizing the text through an embedded model to construct an efficient vector database; secondly, preliminarily retrieving related text blocks by using an improved hybrid retrieval algorithm, and reordering by using a BERT cross encoder to improve the relevance of retrieval results; aiming at the complex task requirement in the automatic driving test, the method is divided into a plurality of sub-tasks to be executed in parallel, so that the retrieval efficiency and accuracy are improved. And then, a LoRA strategy is adopted to finely adjust the large language model so as to adapt to an automatic driving scene generation task. And finally, jointly inputting the cue word template and the reordering result into a fine tuning model to generate a test case. According to the invention, a brand new technical scheme is provided for automatic driving test scene generation, and the method is of great significance in improving the test evaluation level of automatic driving and accelerating the landing of the automatic driving technology.

The invention discloses an automatic driving test case self-evolution repairing method based on a large language model, and belongs to the technical field of testing and performance optimization of an automatic driving system. According to the automatic driving test case self-evolution repairing method based on the large language model, deep semantic mining is conducted on a performance log of the model by means of semantic comprehension and recommendation capacity of the large language model, and a high-quality repairing scene is recommended from a structured test scene library. The method comprises the steps of performance log collection and pre-evaluation execution, failure mode semantic extraction, a scene semantic retrieval mechanism, initial recommendation construction, reflection type semantic optimization, few-sample fine adjustment adaptation and a closed-loop self-evolution mechanism. The method has obvious advantages in the aspects of automatic driving test case recommendation, failure mode recognition, rapid model capability adaptation and system closed-loop self-evolution, and the safety, generalization capability and test maintainability of the automatic driving system can be remarkably improved.

The invention relates to a world model construction method, in particular to a high-fidelity lightweight world model construction method for an end-to-end automatic driving test, which is used for constructing a high-fidelity world model and performing knowledge distillation on the world model to solve the problems of huge parameters, low reasoning efficiency and the like of the world model. Model parameters are reduced on the basis of reserving the world model generation capability, and the reasoning efficiency is improved; a CUDA operator is developed in a user-defined mode for the bottleneck part of world model calculation, video memory allocation is optimized, and the reasoning efficiency of a high-fidelity world model is improved based on a single-device multi-thread scheduling and multi-device cooperative calculation method. According to the method, the high-fidelity lightweight world model for the end-to-end automatic driving test can be constructed, the problems that an existing world model is low in multi-modal information alignment precision, poor in cross-view and cross-frame consistency, low in reasoning efficiency and the like are effectively solved, the confidence coefficient of the end-to-end automatic driving system test process is improved, and the test efficiency of the end-to-end automatic driving system is improved. And the testing efficiency of the end-to-end automatic driving system is greatly improved.

The invention discloses an automatic driving safety key simulation scene generation method based on adversarial generation and co-evolution. The method comprises the following steps: receiving a basic traffic scene described by a natural language, generating an antagonistic element scene containing security threats by using a large language model in combination with a traffic safety knowledge base, and analyzing the antagonistic element scene into an executable scene script; constructing a multi-agent confrontation collaboration diagram based on the meta-scene, and recognizing a key background vehicle through a cross-timing attention mechanism in combination with a time mask and time decay mechanism; and performing disturbance optimization on the key background vehicle trajectory to generate an automatic driving test scene. According to the method, a scientific and systematic solution with engineering operability is provided for safety verification of the automatic driving system when the automatic driving system faces real traffic challenges such as multi-source intervention and dynamic collaborative threat, and the method has wide adaptation capability and important industrial popularization value.

The invention relates to an automatic driving simulation test scene generation method based on a text traffic accident report, and aims to apply a generative language model technology in the field of natural language processing, a code generation technology in the field of program synthesis and a fuzzy test theory in the field of software test. The problem that diversified and standardized high-quality simulation test scenes are lacked in the current automatic driving test field is solved, and the method can generate diversified scene files in the OpenSCENARIO format on the basis of natural languages, and mutates seed scenes through multiple transformation operators so as to efficiently discover defects existing in an automatic driving system.

The invention discloses an automatic driving test scene library construction method based on real traffic data, and relates to the technical field of automatic driving, and the method comprises the following steps: based on a selective sensor fusion framework, dynamically adjusting a fusion strategy of a multi-modal sensor according to a current driving environment, and obtaining corresponding scene elements; performing hierarchical classification on scene elements, constructing a risk assessment model, calculating a comprehensive risk score, and preliminarily dividing risk levels; constructing a rule-based classifier by adopting an association rule mining technology on the basis of results of hierarchical classification and preliminary risk grading, and carrying out risk grading on the scene to be evaluated; and the scene elements and the risk levels are stored in a structured manner, and an automatic driving scene library supporting multi-dimensional query is constructed. According to the method, the characteristics of the traffic scene can be captured more comprehensively, scene elements can be identified more accurately, the risk levels of the scene can be divided scientifically, and the scene library is constructed by combining the scene elements and the risk levels, so that the diversity and pertinence of the test scene are improved.

The invention discloses a self-adaptive adjustment method based on driving behavior prediction, and relates to the technical field related to intelligent driving, and the method comprises the steps: carrying out the driving behavior data collection of a driver, and building a driving behavior data set; external environment data mapped with the driving behavior data set are obtained, and after data cleaning is conducted on the external environment data and the driving behavior data set, driving features and environment features are extracted; inputting into a deep learning prediction model, and executing deep learning prediction model training; and performing driving behavior prediction of the driver through the trained deep learning prediction model, and establishing feedback early warning according to a driving behavior prediction result. The technical problems that in the prior art, a training strategy cannot be dynamically adjusted according to individual behavior characteristics of a driver and environment data, so that the driving safety is not high, the driving behavior prediction precision is insufficient, and the examination passing rate is low are solved. The technical effects of improving the accuracy of driving behavior prediction, the driving safety and the driving test passing rate are achieved.

The invention provides an automatic driving test scene generation method based on real traffic data, and solves the problems of low accident data utilization rate, SIL / HIL test splitting and insufficient boundary coverage in the prior art. Comprising the following steps: acquiring multi-source heterogeneous traffic accident data; cleaning data by adopting a joint interpolation-anomaly detection mechanism; vehicle dynamic sudden change characteristics within 0.5 second before braking are extracted through LSTM and DTW algorithms; constructing a three-dimensional scene pipeline driven by a physical engine, and dynamically associating the pavement slippery coefficient with the rainfall intensity; analyzing the accident text into simulation parameters by using a semantic-physical parameter converter; performing SIL-HIL cooperative verification: performing extreme illumination perception test and narrow road planning verification in an SIL environment, and realizing 1ms step length fault injection test in an HIL environment; positioning failure parameters based on Bayesian optimization; a GAN is adopted to generate a long-tail scene, and a test boundary is expanded by coupling extreme conditions such as rainstorm / low visibility; and outputting a standard scene library containing the collision probability thermodynamic diagram. The safety verification efficiency under the extreme working condition is remarkably improved.

The invention provides a method for testing identification and response of an automatic driving vehicle to a signal lamp in a closed site, and belongs to the technical field of automatic driving testing. The method solves the problem that it is difficult to dynamically simulate a real signal lamp interaction scene and quantitatively evaluate the accuracy of a vehicle perception-decision chain in a closed test. According to the technical scheme, the method comprises the following steps: constructing a line segment topology network with directional attributes; deploying a three-dimensional adjustable signal lamp device at the node to dynamically adjust the pose; synchronously acquiring signal lamp identification result data and control response behavior data of the vehicle; calculating a time margin based on the vehicle positioning speed and the distance from the stop line, and triggering signal lamp state grading switching through comparison with a preset threshold value; a two-dimensional verification mechanism is adopted, and identification result data and control response behavior data are aligned through a time synchronization protocol to analyze causal relevance. The method is used for testing and verifying the recognition accuracy and response timeliness of an automatic driving vehicle on a signal lamp scene in a closed site.

The present invention relates to an autonomous driving test method based on multi-agent swarm adversarial, a device and a medium. The method includes: deducing a conflict topological relationship graph between a tested autonomous vehicle and an agent; deducing a feasible planning space of the tested autonomous vehicle according to the conflict topological relationship graph; establishing a multi-agent swarm adversarial model based on a potential game under the feasible planning space according to a correlation between an individual reward of the agent and a swarm adversarial test effect of a multi-agent system, and solving and obtaining an optimal adversarial strategy of the multi-agent system against the tested autonomous vehicle, where in the multi-agent swarm adversarial model, an adversarial intensity is introduced, and the adversarial intensity is adaptively adjusted according to an actual response of the tested autonomous vehicle; and repeatedly executing the S1-S3 until an adversarial task is completed.

The invention provides a vehicle auxiliary driving test scene generation method and device, equipment and a medium, and relates to the technical field of vehicles, and the method comprises the steps: obtaining the real-time state data and environment data of a vehicle; calculating a dynamic security threshold based on the real-time state data and the environmental data; the dynamic safety threshold value represents the safety risk of vehicle auxiliary driving in a vehicle real-time state dynamic change scene; under the condition that the dynamic safety threshold triggers and adjusts the complexity of the vehicle aided driving scene, guiding a preset test scene generation model to generate a test scene of vehicle aided driving based on the dynamic safety threshold; the test scene is matched with the safety risk. According to the invention, the safety risk can be fed back in real time based on the dynamic safety threshold so as to reduce the braking decision error, thereby improving the accuracy and comprehensiveness of verifying the safety of the auxiliary driving system.

The invention relates to the technical field of automatic driving, in particular to a multi-view-angle perception-driven automatic driving test scene twin reconstruction method, which comprises the following steps of: S1, acquiring and preprocessing multi-view-angle image perception data to form a multi-view-angle image sequence under a unified coordinate system; s2, spatial coding and modeling based on an orthogonal three-plane structure; s3, on the basis of a view perception-space interaction encoder of a Transform architecture, completing space coding processing represented by an orthogonal three-plane structure; s4, structure and semantics combined twinborn reconstruction decoding is carried out; and S5, generating a simulation compatible format and deploying a platform. According to the method, the terrain, facilities, traffic elements and other contents of a real road scene can be automatically restored, the storage and calculation cost is remarkably reduced while the spatial information integrity is ensured, various simulation platform support formats can be flexibly output, the method serves for closed-loop testing of automatic driving, and the reconstruction efficiency and the system suitability are improved.

The invention relates to the technical field of automatic driving test, and discloses an automatic driving network connection cloud control site test method fusing vehicle-road data, comprising the following steps: S1, constructing a dangerous scene set based on roadside sensing data; s2, selecting a test scene; constructing a continuous test task sequence, and solving an optimal solution of the scene risk degree and the coverage degree; executing target object process allocation and global trajectory planning based on the task sequence, and generating an automatic driving continuous test scheme; s3, establishing real-time communication connection between the controlled target object terminal and the cloud; receiving a planning track issued by the cloud at a controlled target object terminal, and constructing a system state model to realize tracking control; a global multi-target task state is monitored at a cloud end, a scene operation cost index is constructed, a multi-target trajectory planning scheme is dynamically adjusted, and a scene task execution state is researched and judged; and S4, constructing a digital twinborn model, and mapping and visualizing test data in real time. According to the invention, an optimization scheme is provided for the high-order intelligent driving function closed site test.

The invention belongs to the technical field of automatic driving test, and provides an automatic driving confrontation test scene generation method and system based on strategy switching, and the method comprises the steps: obtaining an automatic driving simulation scene; setting positions and initial speeds of a target vehicle and an intelligent body vehicle in the acquired simulation scene; in combination with the set position and initial speed of the target vehicle, the running state of the target vehicle is optimized by adopting a multi-agent near-end strategy, and the target vehicle is controlled; determining a current traffic state according to the control driving state of the target vehicle; in combination with the dynamic quantitative index of the artificial risk potential field, sensing the traffic flow risk of the intelligent body vehicle in the determined current traffic state in real time; and according to the obtained traffic flow risk level, automatically switching a defense strategy and an attack strategy of the agent vehicle, and driving the multi-agent vehicle to cooperatively generate an automatic driving confrontation test scene.

The invention discloses an emergency lane changing dangerous scene generation method and system based on behavior guidance and a medium. The method comprises the following steps: constructing a track data set of vehicle emergency lane changing behaviors; training a behavior learning module with short sequence learning ability according to the trajectory data set; training an active attack strategy model through guidance of a behavior learning module; outputting a reference track point of the active attack vehicle through the active attack strategy model, and obtaining an optimal control input in a future time domain according to the reference track point and a preset prediction control model; and controlling the active attack vehicle to interact with the simulation platform according to the optimal control input, and generating an emergency lane changing dangerous scene. The method improves the generation efficiency of the emergency lane changing dangerous scene, thereby improving the efficiency of the automatic driving test, and can be applied to the technical field of automatic driving.

The invention relates to an automatic driving automobile test method based on multi-traffic-participant collaborative hunting confrontation, which comprises the following steps: dividing a test scene into a plurality of space-time resource grids based on scene information, and constructing a conflict event set according to tracks of a tested automobile and traffic participants; defining a conflict degree index, a safety index and passing time; a multi-target speed optimization search model is constructed based on the conflict degree index, the safety index and the passing time, a plurality of traffic participants are utilized to cooperatively surround the tested vehicle, the multi-target speed optimization search model is solved in the cooperative surrounding process, and a space-time resource occupation table is obtained; carrying out collision detection, and if no conflict exists, generating an optimal track of all traffic participants; otherwise, returning to execute collaborative hunting; and obtaining a comprehensive index of the tested vehicle when the traffic participant executes the optimal trajectory, and carrying out performance evaluation. Compared with the prior art, a more accurate automatic driving test is realized by constructing a complex scene considering the collaboration between traffic participants.

The invention discloses a vehicle driving simulation test method and system based on digital twinning, belongs to the technical field of vehicle driving, and realizes real interaction of a driving behavior model and a control algorithm in a virtual scene by constructing a digital twinning body highly corresponding to an actual vehicle and a driving environment. Time stamps and coordinate information of all the modules are collected, time differences and space residual errors are extracted, a space-time semantic consistency index model is constructed, and the loading process is intelligently judged and optimized; when inconsistency is detected, adjustment is performed through clock synchronization and coordinate recalibration, a closed-loop optimization mechanism is formed, and high collaboration of the model and the environment is ensured, so that the accuracy, the stability and the credibility of the simulation system are improved, and the method is suitable for complex and changeable automatic driving test scenes.

The application provides a test method for intelligent auxiliary driving test, constructs a test problem description grabbing model, directly collects signals uploaded by an intelligent vehicle in real time through a vehicle-mounted Ethernet and a CAN FD direct connection interface without using other data interfaces, triggers a test problem description reporting process once abnormal signals are monitored, and ensures that the vehicle meets safety requirements of real vehicle road test.

The invention discloses an automatic driving test method based on traffic regulation weights and electronic equipment, in a test scene generation process, importance weights of traffic regulations are introduced, and the importance weights guide an attention mechanism to distribute different attention degrees to word block vectors, so that the test scene generation efficiency is improved. For example, high attention is allocated to word block vectors corresponding to high-risk behaviors and key regulations, so that subsequent scene generation is ensured to preferentially cover violation scenes of the high-risk behaviors and key traffic regulations; and learning each word block vector and the context thereof in the word block vector sequence through an attention mechanism to obtain global context representation which can reflect global dynamic interaction information in a complex scene, so that diversified and complex test scenes can be generated, and the test efficiency is improved. Therefore, when the test scene generated by the method is used for testing the automatic driving system, the compliance performance of the automatic driving system in a complex and high-risk traffic scene can be reflected more accurately.