126 results about "Traffic simulation" patented technology

The invention relates to the technical field of artificial intelligence and automatic driving, and discloses an automatic driving digital twin agent system based on reinforcement learning training, which comprises an agent library construction module, a simulation and configuration module, a distributed simulation execution module, a scheduling and learning module, a strategy training module and a simulation migration module. The method comprises the following steps: constructing a modular digital twin agent library, combining to generate an urban traffic simulation scene, executing large-scale parallel simulation on a distributed computing cluster, implementing dynamic task scheduling and course learning according to an agent learning progress, and training a reinforcement learning strategy and a value network based on simulation data. And verifying the strategy model and implementing migration deployment from simulation to reality. According to the method, the training efficiency can be improved, the diversity of simulation environments is enriched, and the sim-to-real migration difficulty is reduced, so that the robustness of the strategy model is improved.

The invention discloses an expressway reconstruction and extension traffic diversion method, system and device and a medium, and the method comprises the steps: obtaining multi-source dynamic traffic data, and constructing a road network origin-destination matrix through data fusion and demand deduction; a traffic flow distribution result is obtained by constructing a system optimal traffic distribution model; identifying the first congestion time period, and calculating the traffic flow exceeding the residual traffic capacity in the congestion time period to obtain the traffic flow to be shunted; solving an optimal shunting path under the target of minimizing the total travel time of the system through a hybrid intelligent algorithm to obtain an optimal shunting path scheme; and verifying the optimal shunting path scheme by using microscopic traffic simulation software, updating traffic data based on a simulation result, and carrying out iterative optimization to obtain a differentiated traffic organization scheme. According to the invention, the traffic flow passing efficiency during the highway reconstruction and extension period is effectively improved.

The invention relates to a traffic simulation agent system construction method based on a large model, and the method comprises the steps: constructing a simulation tool library, and achieving the precise evaluation and continuous optimization of a simulation result through the fusion of multi-source heterogeneous traffic data, the construction of standardized input, and the establishment of a quantitative evaluation system. A large language model is utilized to understand a natural language instruction of a user, tasks are intelligently disassembled, an execution process is planned, dependence management and parallel scheduling are carried out in combination with a directed acyclic graph, and professional tools are driven to automatically execute. And performing evaluation, problem diagnosis and adaptive re-planning on an execution result through a large language model reflection mechanism to form an understanding-planning-execution-reflection closed loop. According to the method, the problems of how to assist a user to interact with a traffic system by utilizing an agent technology driven by a large language model, reducing the technical threshold of traffic simulation software use and saving time cost and labor cost are solved, the traffic simulation use threshold is reduced, the automation and intelligence level is improved, and efficient and accurate traffic system interaction and optimization are realized.

The invention discloses an intelligent decision-making system construction method for traffic signal control. The method comprises a model training and deployment stage and an application and evolution stage, and specifically comprises the following steps of: S1, generating a pairing training sample set of traffic state structured data and conflict-free signal control instructions based on a pre-stored road traffic conflict rule in the model training and deployment stage; utilizing the paired training sample set to supervise and finely adjust a large language model to obtain a basic model; s2, accessing the basic model into a traffic simulation environment for reinforcement learning training; and in each training step, the basic model outputs a signal control instruction according to the current traffic state, performs safety verification on the instruction according to the road traffic conflict rule, generates a safety reward signal and the like. The traffic signal intelligent decision-making system which is safe, credible, sustainable in evolution and suitable for edge independent deployment is constructed.

The invention discloses an encrypted domain name resolution protocol simulation and characterization system, and relates to the technical field of network security and network traffic analysis. The invention aims to simulate an encrypted domain name resolution process in a real network environment, collect the flow of the process and extract features to construct a data set, and ensure the quality of the generated data set through data enhancement and a data set evaluation scheme. The system comprises a traffic simulation module, a traffic representation module, a data enhancement module and a data set evaluation module. The flow simulation and characterization module simulates and encrypts domain name resolution flow and extracts a structured feature vector containing 34 side channel features; and the data enhancement and data set evaluation module is used for enhancing a feature set based on a conditional table generative adversarial network CTGAN so as to construct a feature data set which is closer to traffic in a real network environment, and ensuring that the constructed data set has engineering availability and theoretical rationality through evaluation. The system can be used for constructing a current scarce encrypted domain name resolution protocol side channel feature data set, and provides data support for related security detection and research.

The invention relates to the technical field of intelligent traffic, in particular to an optimization method and device for a traffic control scheme generation model, equipment and a medium. The method comprises the steps of obtaining a traffic event training data set; inputting each traffic event into a traffic control scheme generation model, and generating at least one candidate control scheme of the traffic event; sequentially inputting the at least one candidate control scheme into a traffic simulation model for deduction to obtain parameters of at least one simulation index of each candidate control scheme; calculating a reward value of each candidate control scheme based on each candidate control scheme, the parameter of the at least one simulation index and a reward function; wherein the reward function comprises a format reward function, a rule reward function and a quality reward function; and based on the reward value of each candidate control scheme and the real annotation data of the traffic event, optimizing the traffic control scheme generation model by adopting a reinforcement learning algorithm based on strategy gradient to obtain an optimized traffic control scheme generation model.

The invention relates to a semantic-driven traffic simulation scene generation method based on a large language model. The method comprises the following steps: receiving traffic simulation scene natural language description input by a user; constructing a tool knowledge base, performing retrieval in the tool knowledge base based on user input, and combining a retrieval result with the user input to obtain extended input; calling the large language model to carry out multi-step reasoning on the extended input, determining a tool needing to be called in a tool knowledge base, and generating a structured simulation configuration element by utilizing the determined tool; and converting the generated simulation configuration elements into a configuration file which can be directly loaded to a traffic simulation platform, and executing simulation verification in the traffic simulation platform. Compared with the prior art, natural language input can be automatically converted into a structured and executable traffic simulation configuration file under the condition that manual intervention is not needed, and high efficiency and intelligence of scene construction are achieved.

The invention discloses an intelligent vehicle autonomous lane changing decision-making system and method based on near-end strategy optimization. Belongs to the technical field of intelligent traffic and automatic driving. The technical problem that in the prior art, defects still exist in the aspects of state space design, security constraint modeling, training strategy optimization and the like is solved. The system comprises a traffic simulation environment module, a state observation and feature construction module, a near-end strategy optimization lane change decision module, a reward calculation and safety supervision module, a training control and data recording module and a strategy export and application module. A training environment is constructed based on a microscopic traffic simulation platform, a training scene is closer to real road traffic by configuring different traffic flow densities and speed distributions, and the strategy has better adaptability under different working conditions.

The invention relates to the technical field of intelligent traffic simulation and heavy vehicle cooperative control, and discloses an intelligent cooperative control simulation verification system for a tractor or a trailer, which comprises three modules and four modules: a three-dimensional environment modeling module based on multi-source data fusion and hierarchical modeling and combined with a timing and event double-trigger mechanism to update dynamic obstacles; the cooperative motion control module integrates Bezier trajectory planning, a geodesic theoretical dynamics coupling model and triangular collision detection, and realizes vehicle-towing cooperation and risk early warning. The visual simulation interaction module constructs a multi-window view based on OpenGL / Unity 3D, and simulates light and shadow and friction characteristics in combination with material attributes; the result verification module outputs a conclusion based on the quantitative index and the qualitative report. According to the system, the goodness of fit between a simulation environment and a real scene is remarkably improved, the collision early warning response is less than or equal to 100ms, more than 80% of roll-on-roll-off ship scenes can be covered, the entity test cost is reduced by 60%, and a high-fidelity verification tool is provided for vehicle-tow intelligent cooperation.

The invention provides an automatic driving formation hardware-in-the-loop simulation platform based on a satellite position simulation platform and a medium, and relates to the technical field of automatic driving and car networking simulation. The platform comprises a satellite position simulation module, a traffic simulation platform, a hardware verification platform, a simulation bus and a clock synchronization module. The satellite position simulation module generates satellite data with a timestamp; the traffic simulation platform constructs a formation driving scene and operates a formation control algorithm; the hardware verification platform comprises a real phased-array antenna module and a PC5 communication device. The modules exchange data through a simulation bus, and the clock synchronization module unifies the time reference to form a closed-loop verification environment of simulation hardware simulation. The problems that space-based data and ground simulation are separated and real communication characteristics are difficult to quantitatively evaluate in the prior art are solved, the satellite-ground cooperation and formation control strategy can be efficiently and reliably verified in a laboratory environment, and the real vehicle test cost and risk are remarkably reduced.

The invention relates to the technical field of urban traffic management, in particular to a traffic simulation data generation method and system based on a data distillation and knowledge distillation technology, and the method comprises the steps: obtaining original data of a real traffic scene, carrying out the cleaning, standardization and spatial-temporal feature extraction of the obtained original data, and obtaining a structured feature set; performing data distillation analysis on the structured feature set to obtain an initial simulation data set conforming to real distribution; inputting the initial simulation data set into a knowledge distillation model, and outputting the knowledge distillation model to obtain knowledge distillation parameters; performing fusion processing on the knowledge distillation parameters and the initial simulation data set to obtain a fusion enhanced data set; and performing dynamic scene adaptability verification and multi-dimensional distribution correction on the fusion enhanced data set to obtain a final simulation data set.

A system is provided.SOLUTION: A system comprising: means for real-time analysis of traffic information collected from cameras and sensors; means for generating a traffic simulation using a generative AI model; means for calculating optimal traffic routes and signal light timing; means for sending instructions to signal lights and digital signage based on the calculation results; means for continuously monitoring and optimizing traffic conditions; and means for notifying users of traffic information.SELECTED DRAWING: Figure 1

This invention discloses a system for identifying and simulating key nodes in traffic networks based on deep reinforcement learning. Addressing the high computational complexity of key node search and the lack of dynamic verification in traffic networks, this invention employs the following system: a road network topology mapping module parses GIS map data and removes redundant information to generate a weighted topology map; a deep reinforcement learning identification module, based on a deep Q-network, maps node features to graph embedding information through graph representation learning, with the optimization objective of minimizing cumulative normalized connectivity, outputting the optimal key node sequence; and a traffic effect simulation verification module uses a traffic simulation platform to establish a realistic road network model, quantitatively analyzing the impact of key node failures on average travel time and waiting time. This invention features low time complexity, the ability to generalize from small synthetic networks to extremely large-scale real networks, and simulations have verified its high application value in actual traffic management and disaster prevention.

The invention discloses a traffic scene simulation method based on an accident report, and aims to solve the problem of insufficient complex dynamic scene data in traffic simulation. According to the method, physical clues (such as vehicle speed, collision angle, quality and the like) and context information (such as road types and environmental conditions) are extracted from a real traffic accident report, and a high-fidelity traffic scene data set is generated. In foreground processing, the extracted physical clues are used for accurately constructing a vehicle motion trajectory, diversified dynamic traffic scenes are generated through a pre-collision trajectory planning algorithm, and trajectory prediction is optimized in combination with a large language model to ensure that the real physical law is met. In background processing, high-quality three-dimensional scene reconstruction is carried out according to environment information in an accident report, a vivid background image is generated, and the consistency of a visual angle and illumination is kept. And finally, through fusion of the foreground vehicle track and the background image, a visual and physical double real traffic simulation video is generated. The method significantly improves the diversity and authenticity of traffic scenes, reduces the dependence on real data collection, is widely suitable for high-fidelity traffic simulation scenes such as automatic driving test and traffic flow analysis, and effectively promotes the progress of the traffic simulation technology.

The application relates to a semantic driving traffic simulation scene generation method based on a large language model, which comprises the following steps: receiving a traffic simulation scene natural language description input by a user; constructing a tool knowledge base, searching in the tool knowledge base based on the user input, combining the search result with the user input to obtain an extended input; calling a large language model to perform multi-step reasoning on the extended input, determining tools required to be called in the tool knowledge base, and generating structured simulation configuration elements by using the determined tools; converting the generated simulation configuration elements into a configuration file which can be directly loaded into a traffic simulation platform, and performing simulation verification in the traffic simulation platform. Compared with the prior art, the application can automatically convert natural language input into structured and executable traffic simulation configuration files without manual intervention, and realizes the efficiency and intelligence of scene construction.

This invention discloses a collaborative control method for highway merging zones based on deep reinforcement learning. A LiikeSim-Python co-simulation environment is established, and loop detectors are set up in the simulation environment to acquire traffic flow data upstream and downstream of the highway merging zone. An EM algorithm based on Gaussian mixture distribution is used as a traffic state classifier, taking the traffic flow data of the highway merging zone as input to classify the traffic state of the merging zone. A state space, action space, and reward function are designed. Using the state space of the highway merging zone as input and the actions of the variable speed limit agent and the ramp metering agent as output, a multi-agent shared experience network model under time-series characteristics is constructed. An independent experience pool is set up for the variable speed limit agent and the ramp metering agent, and the interaction experience between the agent and the traffic simulation environment is collected at the control cycle frequency. The agent model is trained using sampled data. The trained agent model is used to realize collaborative control of the highway merging zone. This invention can reduce travel delays in highway merging zones.

This invention discloses a method and system for generating traffic simulation scenarios based on a large language model, belonging to the field of autonomous driving technology. The method includes: parsing the natural language description of the scene into a structured descriptor, wherein the structured descriptor includes: static scene information, scene elements, scene element behavior information, and environmental information; describing the static scene based on the static scene information, scene elements, and environmental information to generate a static initial scene; reasoning about the scene elements in the static initial scene based on the scene element behavior information to generate a control chain that conforms to semantic logic, satisfies physical execution constraints, and aligns the timing of scene element actions; and generating a traffic simulation scene based on this control chain. This invention can improve the flexibility, richness, and generalization of scene generation.

The invention relates to a method for simulating traffic, having the steps of: providing (S1) a data set with information on a real traffic scenario using movement data of traffic participants, beginning (S2) the process of running the traffic simulation with simulated counterparts of the real traffic participants using movement data according to the data set with the exception of the movement data of one vehicle of the real traffic participants to be replaced, as a virtual test vehicle, repeatedly checking (S3) in a sequential manner whether a metric relating to a deviation of the movement data of a respective non-replaced traffic participant according to the data set from the planned movement data calculated for the traffic participant in the simulation exceeds a specified threshold and maintaining the movement data according to the data set if the specified threshold has not been exceeded, otherwise: replacing / mixing (S4) the movement data taken on according to the data set with movement data calculated by an agent model and carrying out a return trajectory to the movement data according to the data set and ending the process of running the corresponding agent model in the event of a specified termination condition.

Provided is a traffic simulation for controlling a motion of an object, such as a vehicle, a pedestrian moving on a road or a pavement, in a driving simulation, an autonomous driving simulation, or the like. A traffic simulation method according to an embodiment of the present disclosure includes the steps of: importing a new moving object into a simulation environment of a simulator; retrieving data of a moving path and a start point of the moving object which is created based on a function, among pre-stored data; calculating 3D coordinates regarding a position of the moving object; moving the moving object along the moving path in the simulation environment, based on the calculated 3D coordinates; and calculating a next position of the moving object.

The application provides a charging pile layout optimization method, system and equipment, relates to the technical field of charging facility management, and comprises the following steps: acquiring historical traffic flow and power grid node capacity, generating a charging pile coordinate set and determining involved road sections; wherein, the involved road sections are the necessary road sections of the navigation path within the influence range of each charging pile; potential congestion road sections in each period are identified by using a traffic simulation model; if there is congestion, the total number of stranded vehicles is calculated, a time-varying charging load curve is generated in combination with the electric vehicle proportion and the single-vehicle fast-charging power; the curve is mapped to the power grid node, whether the node voltage is lower than a threshold value is evaluated, and the pile-to-be-supplemented road section is confirmed; a candidate coordinate set is generated around the pile-to-be-supplemented road section, the supplementary charging pile position coordinates are screened out and are incorporated into the original coordinate set, and a complete layout scheme is formed. The application realizes the optimization of the charging pile layout, improves the utilization rate of the charging pile, and ensures the stable operation of the power grid.

The application discloses a kind of microcosmic traffic simulation parameter calibration value data processing method of group difference minimum, it is related to traffic safety and traffic simulation technical field.The method is realized by accurate measurement and fine operation to simulation model parameter, and the optimization of microcosmic traffic simulation parameter.Firstly, by setting traffic volume, speed, vehicle type composition, running time and other parameters, using VISSIM simulation model and MATLAB realizes simulation calibration program.Then, the program is run, and the parameter calibration result is obtained, and the average value and standard deviation of each parameter are calculated.Subsequently, the difference between the parameter and the average value is calculated, and the size of the difference and the standard deviation is compared, and the data that does not meet the condition is eliminated.Finally, the group data that the error of each parameter is within the standard deviation range is retained, the difference value of each parameter is obtained, then the data is arranged, and the data difference minimum is taken out as the final parameter value result.The method can greatly improve the accuracy and reliability of simulation model.

This invention relates to the field of traffic simulation and pedestrian flow modeling technology, and particularly to a method and system for crowd movement modeling based on a large language model. The method includes: constructing a multi-dimensional interactive feature system of pedestrians, small groups, and the environment; extracting micro-behavioral features of small pedestrian groups through micro-unit segmentation and dynamic group partitioning methods; proposing a pedestrian movement decision-making framework based on a large language model and thought chain; designing a hybrid update decision-making strategy to balance high-level semantic decision-making and local obstacle avoidance; and training and testing pedestrian evacuation at traffic hubs. Instance verification of pedestrian small group movement decision-making at traffic hubs is also conducted. The proposed model is applied to hub scenarios and its empirical dataset for training and testing. Results show that it achieves better motion realism, traffic efficiency, and group structure stability than baseline models.

The application discloses a kind of multi-scale traffic facilities three-dimensional model high-efficiency rendering method and system, it is related to computer graphics and three-dimensional visualization technical field.For the problems that traffic three-dimensional scene static LOD rendering is prone to picture jump, rendering load fluctuation is violent, traffic key semantic simplification loss, the application is first based on camera motion parameter and constructs dynamic three-level sector partition;Then through lightweight CNN extraction semantic mask, fuse semantic weight constraint edge folding simplification, 100-300ms interpolation is realized to smooth LOD switching;Finally, construct quantization rendering cost model, rely on double-threshold closed-loop control dynamic adjustment model simplification degree.The application gives consideration to model light weight and traffic feature fidelity, stable rendering frame rate, and can be widely applied to urban traffic simulation, road digital twin, vehicle-mounted real scene visualization and the like.

The application discloses a kind of microcosmic traffic simulation data's map engine adaptation conversion method and system, it is related to traffic simulation technical field.The method includes: extracting road section information and lane information from simulation road network file;According to road section information, judge road section nature, according to road section information and road section nature, construct independent attribute set for each road;Based on driving lane line shape and lane width, generate driving lane surface file for each road, the independent attribute set constructed is used as the attribute of corresponding driving lane surface file;Parallel offset is carried out to driving lane line shape, form emergency lane surface file;Driving lane surface file and emergency lane surface file are carried out gap correction;According to independent attribute set, establish style rule library, judge the required lane type, realize automatic marking line adaptation according to lane type, generate lane line file;Lane line file, corrected driving lane surface file and emergency lane surface file are loaded to geographic information service software.

The application relates to a kind of intersection full-link traffic simulation methods based on discrete grid structure, comprising the following steps: first, build the traffic simulation environment based on deep discrete grid structure, comprising: determining simulation time accuracy and discrete grid accuracy;Intersection space discretization processing;Define intersection space grid attribute;Determine data storage mode;Build vehicle discretization mapping and driving model, traffic individual generation model;Afterwards, establish intersection game decision model and trajectory planning model respectively, the game decision result of game decision model output is used as the prior input and constraint condition of trajectory planning model, and the trajectory planning and execution result output by the trajectory planning model is used as the input of game decision model, to build the full-link simulation process of game interaction decision-trajectory planning. Compared with the prior art, the application can realize high-precision simulation of complex environment and truly reproduce irregular motion of traffic individuals.

The application relates to a smart container truck platoon driving simulation platform, which comprises three modules of traffic simulation, platoon management and truck control connected in sequence, and the traffic simulation module is interactively connected with the platoon management module and the truck control module. The traffic simulation module is used for human-driven vehicle simulation, and the module is composed of seven units of road network, path decision, traffic control, traffic detection, human-driven vehicle control, information visualization and traffic generation; the platoon management module is used for intelligent container truck platoon behavior simulation, and the module is composed of two units of fleet decision maker and platoon management strategy; and the truck control module is used for intelligent container truck simulation, and the module is composed of three units of single vehicle decision maker, controller and vehicle dynamics. Compared with the prior art, the application supports verification of various platoon driving functions, can evaluate the influence of intelligent container truck platoon on the traffic system under the background of new mixed traffic flow, and can evaluate the efficiency of the platoon management strategy.

The invention discloses a ship lock full-automatic motion control method based on traffic simulation and secondary scheduling fusion, and the method comprises the steps: constructing a data-driven intelligent decision closed loop, and achieving the full-process automatic scheduling of a ship lock through multi-source data perception, information fusion processing, intelligent decision generation, execution control and operation feedback. Comprising the following core steps: establishing a ship scheduling state machine model to realize ordered migration of a ship lockage state; constructing a pre-scheduling mechanism, and establishing a trigger condition and a rolling retention mechanism by combining the approach channel bearing capacity model and the ship comprehensive time estimation model; compared with the prior art, the method has the advantages that a five-layer closed-loop architecture is constructed, and the ship dispatching state machine model containing six main states is established, so that the whole process of ship registration, queuing, marshalling, gate entering and gate leaving is automated; manual intervention links are reduced, manual errors caused by manual scheduling are reduced, the operation cost input is reduced, and meanwhile the standardization degree and the overall efficiency of scheduling operation are greatly improved.

The application discloses a vehicle state management method and device based on a simulation environment, equipment and products, and belongs to the technical field of simulation. The method comprises the following steps: acquiring a traffic simulation environment, wherein the traffic simulation environment comprises a target lane; a plurality of vehicle queues connected in front and back of the target lane are maintained, wherein the vehicles in each vehicle queue correspond to the same type of motion state, and the motion states corresponding to two adjacent vehicle queues in front and back are different, and the motion state comprises a forward state and a stop state; based on the passing state corresponding to the target lane, target vehicles in the plurality of vehicle queues that meet the updating condition are determined; and an updating operation is performed on the target vehicles, and the updating operation is used for updating at least one of the position, speed and motion state of the target vehicles, wherein the stationary vehicles in the stop queue except the passable vehicles are skipped, the stop queue refers to the vehicle queue in the stop state, invalid calculation can be eliminated, and the vehicle state updating efficiency of the simulation frame is improved.

The invention discloses an automatic driving vehicle performance boundary mining method, system and device and a medium. The method comprises the following steps: decoupling five elements of people, vehicles, roads, environments and tasks representing an automatic driving traffic simulation test scene; performing parameterized expression on the five types of elements; screening out a key parameter set of five types of related elements; taking the screened key parameter set as a decision variable, taking an automatic driving simulation platform as an evaluation tool, and taking the minimum speed difference absolute value and distance between a tested automatic driving vehicle and surrounding dynamic traffic participants and static objects as optimization targets, and establishing a double-target random simulation optimization model for automatic driving performance boundary mining; and solving by adopting a double-target monkey group random agent optimization algorithm to obtain a Pareto random optimal key parameter solution set so as to obtain the performance boundary of the tested automatic driving vehicle. According to the method, the performance boundary of the automatic driving automobile is accurately mined by combining the virtual simulation test and the proxy optimization technology.

The invention provides a multi-agent traffic signal control method based on hierarchical comparative learning, and the method comprises the steps: constructing a traffic simulation environment according to a traffic network, and modeling each intersection in the traffic simulation environment as an independent agent; the intelligent agent interacts with the SUMO simulation environment, collects traffic state information, signal control actions, instant rewards and traffic state information of the next state of each intersection, and stores the traffic state information, the signal control actions, the instant rewards and the traffic state information of the next state into an experience playback buffer area; the method comprises the following steps of: grouping intelligent agents to generate regional pseudo labels; according to different region pseudo labels, region division is carried out, sub-graphs are constructed, intra-region feature aggregation is carried out on the sub-graphs, and refined credit distribution of structure perception is realized; and finally, joint optimization is carried out based on comparative learning and a QTRAN framework, so that efficient multi-agent cooperative control is realized. According to the invention, an optimal signal timing strategy can be provided for each intersection, the road network traffic efficiency is improved, and the vehicle queuing length and waiting time are significantly reduced.