501 results about "Traffic efficiency" patented technology

The invention belongs to the field of artificial intelligence and intelligent traffic systems, particularly relates to a vehicle path planning system based on multi-agent collaboration, and aims to solve the problems of path response lag, insufficient global optimization ability and conflict between individuals and group targets in a dynamic traffic environment. The system adopts a three-level collaborative architecture of a central coordination server, a regional scheduling agent and a vehicle-mounted agent, realizes macroscopic regulation and local resource allocation through a global guidance potential field and an improved auction algorithm, and improves the dynamic adaptive capacity by combining event-driven replanning and a robustness verification mechanism. Different scheduling and multi-time scale coupling operation of heterogeneous vehicles are supported, and the road network passing efficiency and the path reliability are effectively improved.

The invention is suitable for the technical field of automatic driving, and provides an automatic driving vehicle blind area track prediction and planning method based on Transform, and the method comprises the steps: firstly constructing a key entrance point, calculating a shielding region, and then generating and maintaining a belief state of a ghost target in combination with a risk score; then, real and ghost targets are coded into Token in a unified mode, the Token is input into Transform for interactive modeling, and multi-modal track distribution is output; and finally, a variable anisotropic repulsive force potential field is constructed based on the predicted risk, and a safety control instruction is obtained through optimization solution and hard safety filtering. According to the method, continuous probabilistic estimation and interactive reasoning of the potential risk of the blind area are realized, a flexible defensive track can be generated according to the risk confidence and the dynamic state of the vehicle, and the traffic efficiency is guaranteed while the safety is improved.

The invention relates to the field of intelligent traffic control, discloses an intelligent control method and system for a bidirectional driving wheel type sightseeing train, and aims to solve the problems of low efficiency, poor safety, weak adaptability and the like caused by dependence on a fixed track, path solidification and control lag of a traditional sightseeing train. According to the method, a dynamic environment map is constructed by deploying a front-back symmetric multi-mode sensor array and fusing laser radar, vision, millimeter wave radar and inertial data; an improved mixed A star algorithm is adopted to plan a global collision-free trajectory, high-precision tracking is realized in combination with model prediction control, bidirectional autonomous operation, multi-vehicle intelligent cooperation and full-scene adaptive control under a trackless condition are realized, and the traffic efficiency, safety and intelligent level of a scenic spot are remarkably improved.

The invention provides a building interior design layout optimization method, and relates to the technical field of indoor modeling and layout optimization, and the method comprises the steps: obtaining building structure boundary data and current situation layout information, constructing an indoor three-dimensional space model, and completing the marking of node passing attributes and a spatial topological structure. On the basis, crowd behavior data of a daily scene and an emergency scene are collected, behavior characteristics such as walking speed, obstacle avoidance response and congestion sensitivity are extracted, and a walking characteristic model used for simulating the individual motion trend is established. And performing multi-scene simulation under a daily event and an emergency evacuation event by using the walking feature model, calculating performance indexes such as passing time, local congestion degree, key node congestion probability and total evacuation time, and identifying a congestion sensitive area. And further constructing a multi-target evaluation model including traffic efficiency, congestion degree and evacuation performance, and performing global optimization on layout adjustable parameters by adopting an algorithm to obtain a candidate layout scheme set.

The invention discloses an urban elevated expressway congestion relieving method and system based on detector data. The method comprises the steps that multi-dimensional traffic parameters of key sections of an elevated expressway are collected in real time through arranged traffic detectors; establishing a multi-level threshold matrix based on the multi-dimensional traffic parameters, and dynamically dividing traffic congestion levels; when the real-time traffic parameters meet the threshold condition of any congestion level, a disposal instruction of the corresponding level is automatically triggered; according to the processing instruction, executing a multi-device cooperative control strategy covering a congestion point space correlation area; and dynamically optimizing the threshold matrix or the cooperative control strategy according to traffic parameter change feedback after strategy execution. According to the invention, by monitoring the traffic data in real time, dynamically setting the congestion threshold and automatically matching the optimal mitigation strategy, rapid early warning and accurate intervention of traffic congestion can be realized, and the traffic efficiency and intelligent management level of the elevated expressway are improved.

The invention is suitable for the technical field of traffic control, and provides a road section pedestrian crossing signal control method and system based on human and vehicle postures. The method comprises the following steps: acquiring instantaneous speed and position data of a head vehicle on each lane in front of a pedestrian crossing in real time, and calculating the maximum crossing time by combining the length of the pedestrian crossing and the type of a pedestrian; determining the length of the upstream and downstream judgment area of the pedestrian crossing based on the road design speed and the street crossing time; monitoring the number of pedestrians and the longest waiting time in the waiting area, and judging whether a first triggering condition is met or not by combining a reasonable waiting time threshold value and a capacity threshold value; judging whether a second triggering condition is met or not by comparing the theoretical driving time of the vehicle with the pedestrian crossing time in the judgment area; and when the first and second trigger conditions are met simultaneously, starting a pedestrian green light phase, otherwise, maintaining the current signal state. According to the invention, the waiting time of pedestrians and vehicles can be effectively reduced, and the intersection passing efficiency and safety are improved.

The invention belongs to the field of intelligent traffic systems, and particularly relates to a congestion scene emergency decision-making method based on dynamic traffic environment modeling, which comprises the following steps: by introducing a congestion potential field concept, fusing multiple factors such as traffic density, speed, cart proportion and the like into a quantifiable risk index, and distinguishing a current congestion index from a future congestion index; the layout problem of the monitoring points is converted into a mathematical optimization problem with the aim of improving the decision effect from experience design; in the decision-making process, not only is passing efficiency considered, but also dimensions such as safety, user experience and resource utilization rate are included; according to the closed-loop intelligent management system integrating traffic flow long-time-sequence evolution prediction, congestion risk assessment based on the physical potential field theory, multi-dimensional comprehensive decision and monitoring point layout collaborative optimization, the opening and closing opportunity of an emergency lane is scientifically and prospectively determined, and the layout of sensing equipment is synchronously optimized; the traffic jam is relieved to the maximum extent, the road passing efficiency and safety are improved, and meanwhile energy consumption is reduced.

The invention discloses an automatic driving decision-making method and system for a port mixed traffic intersection based on deep reinforcement learning, and relates to the technical field of automatic driving. Comprising the following steps: acquiring real-time state information of automatic driving transportation equipment in a port mixed traffic intersection environment, and fusing the real-time state information into an automatic driving transportation equipment state vector in a heterogeneous manner; constructing a reinforcement learning decision model based on the DDPG; building a port traffic environment simulation platform; the reinforcement learning decision model based on the DDPG is trained; and solidifying and deploying the trained reinforcement learning decision model based on the DDPG in a vehicle-mounted calculation unit of the automatic driving transportation equipment, and controlling the driving of the automatic driving transportation equipment. The traffic efficiency, stability and safety of the automatic driving transportation equipment at the port intersection can be improved, the unmanned transportation process of the port mixed traffic scene is promoted, and the wharf transformation and operation cost is reduced.

The invention provides a robot cluster collaborative collision avoidance method and device for electric power system inspection, and the method comprises the following steps: building a maneuvering library for each robot in a robot cluster, and enabling each robot to broadcast a data package containing the motion state of itself and a planned path in the cluster; based on the data packet, identifying robots with potential collision conflicts, and enabling the robots to form a temporary decision group; in the decision group, each conflict robot determines an intention proposal, and calculates a negotiation priority according to the task priority, the urgency degree of the potential collision conflict and the historical maneuvering frequency; and according to a preset decision and verification rule, allocating a conflict-free collaborative collision avoidance action to each conflict robot. By implementing the technical scheme provided by the invention, the congestion risk caused by the fixed priority is effectively solved, and the passing efficiency of the whole robot cluster is remarkably improved.

The invention discloses an inland waterway congestion dredging method and device, and relates to the field of traffic control, and the method comprises the steps: obtaining multi-source traffic data and waterway meteorological and hydrological data of each ship in a waterway; performing space-time registration and fusion processing on the multi-source traffic data, generating comprehensive traffic situation information of the whole section of the channel, and extracting traffic flow characteristic parameters and channel congestion indexes; inputting the traffic flow characteristic parameters, the channel congestion index and the channel meteorological and hydrological data into a trained traffic flow prediction model to obtain traffic prediction information of key nodes in the channel in a future preset time window; based on the comprehensive traffic situation information and the traffic prediction information, generating preventive control parameters for channel key node signal lamps; and according to the traffic prediction information and the preventive control parameters, traffic guidance information is generated and sent to the corresponding ship terminals, and the preventive control parameters are sent to the corresponding signal lamp control machines. The overall passing efficiency of the channel can be improved.

The invention discloses an evolutionary game theory-based overtaking and lane changing behavior control method for an expressway interworking and interleaving area, and belongs to the technical field of intelligent network connection vehicle cooperative control. According to the method, firstly, a lane changing opportunity model is constructed based on vehicle motion state constraints and driving comfort conditions, and all lane changing opportunity points are obtained; through an evolutionary game model, the efficiency, the safety effectiveness and the traffic environment influence of the overtaking vehicle and the overtaken vehicle are integrated, and the optimal lane changing opportunity is screened; and finally, establishing a trajectory prediction model by utilizing model prediction control, and generating an optimal lane changing trajectory. According to the method, the integrated optimization of the lane changing opportunity and the track is realized, the speed fluctuation can be effectively inhibited, the longitudinal risk of the lane is reduced, the lane changing success rate, the passing efficiency and the driving safety of the expressway intercommunication interweaving area are remarkably improved, and the method is suitable for the lane changing control of various intelligent network connection vehicles in a complex interweaving environment, and has a relatively high engineering application value.

The invention discloses an intelligent highway abnormal event identification method based on multi-mode perception, and relates to the field of congestion identification, and the method comprises the steps: collecting the dynamic data of a highway in real time through a sensor network; feature extraction is carried out based on historical and real-time traffic flow data of each road section of the expressway, the road congestion condition is judged through feature fusion, and a congestion index is generated; based on the congestion event, triggering traceability analysis, and generating a feature vector of the source area; carrying out model training through a deep reinforcement learning model, obtaining a trained congestion reason identification and classification model, and obtaining a congestion reason label; and generating early warning information based on an identification result, and updating a navigation route through the linkage navigation platform. The method has the advantages that abnormal events on the expressway are accurately recognized by fusing video monitoring, radar, weather and toll station data, real-time traffic early warning is provided through the deep learning and prediction model, and road safety and traffic efficiency are improved.

The invention discloses a deep reinforcement learning highway variable speed limit control method based on multi-target dynamic weight distribution, and the method comprises the steps: 1, collecting the traffic flow information of a road section, and obtaining the traffic state data; 2, defining an action parameter and a reward function of the deep reinforcement learning method; 3, constructing a network model of deep reinforcement learning, and establishing a playback memory bank storage sample; and 4, reward function weight updating is carried out based on an entropy weight method, an optimized control strategy is obtained based on a sampling training deep reinforcement learning network model, and speed limit control is carried out on each control unit. According to the method, dynamic weight distribution of the traffic efficiency and the safety reward function is achieved, the method can adapt to a complex traffic environment, meanwhile, the operation efficiency of the expressway is effectively improved, and the safety risk is reduced.

The invention belongs to the technical field of expressway passenger and truck lanes, and particularly relates to an expressway passenger and truck lane dynamic allocation optimization method which comprises the following specific steps: S1, data acquisition and preprocessing: integrating ETC and radar multi-source data, cleaning and fusing original data, and accurately distinguishing small passenger cars and trucks by using an advanced vehicle type recognition technology, so as to obtain a new vehicle type recognition technology; a reliable data basis is provided for subsequent analysis; s2, traffic state perception and prediction: calculating key traffic parameters of traffic flow, speed and density based on the preprocessed data; and a scientific prediction model is called. According to the method, the future short-time passenger and freight predicted flow provided by traffic state perception and prediction is received, and the lane allocation decision is made based on the dynamic data, so that the number of the passenger car special lanes can be increased in time when the flow of the passenger cars is sharply increased in the morning peak, the delay of the passenger cars on the passenger and freight mixed lanes is reduced, and the overall traffic efficiency is effectively improved.

The invention discloses a traffic abnormal event detection method based on multi-source data fusion, and the method comprises the steps: obtaining target detection data through a single millimeter wave radar and a plurality of industrial cameras according to the characteristics of a sensor; obtaining fusion information by associating the multi-source information; tracking the fusion target to form a target trajectory, and performing smooth filtering on the trajectory to form a filtered target trajectory; violation event detection is realized by analyzing a target track, wherein violation events comprise invasion, overspeed, underspeed, lane change, illegal parking, reverse driving and line pressing; traffic state event detection is achieved by analyzing the correlation between the lane-level detection area and target data, traffic state events include congestion, event time is aligned for event evidence leaving, and event videos and event photos are included; and for a specific event, event recheck is carried out through target track anti-shake verification, and event qualification and reporting are carried out. The traffic safety risk can be effectively and intelligently prevented and controlled, the traffic efficiency is optimized, and the traffic accident occurrence probability is reduced.

The invention provides a normalized lane charging system and charging method for a highway toll station, and relates to the technical field of intelligent traffic. According to the method, a system lane front-end function module is arranged at the front end of a lane of a toll station and used for directly interacting with vehicles in the lane, collecting vehicle information and executing a control instruction; the edge node processing module is arranged at an edge node of a toll station level, and is used for processing transaction data and picture information from the lane front-end function module based on a virtualization technology, and carrying out collaborative management on multiple pieces of lane service data under the condition that the edge node processing module is matched with the cloud service management module; and the cloud service management module is deployed at a cloud end and is used for carrying out unified management on the charging services of the plurality of toll stations. The system can decouple business processing and equipment control by adopting a cloud-side-end system architecture, and improves the lane passing efficiency.

The invention provides a park emergency navigation edge cooperation system and method based on building structure situation awareness, and relates to the technical field of path planning. The safety situation of the building structure is quantified for the first time and serves as a core optimization target which is the same as the passing time to be fused into path planning, personnel are prevented from being guided to a high-risk area of the structure fundamentally, and the essential safety of evacuation is remarkably improved. By exchanging decision factors among edge nodes and executing distributed multi-target planning, cross-regional path real-time collaboration under centerless scheduling is realized, coordination of a global path network in safety and efficiency is guaranteed, and robustness and autonomous decision ability of a system when a local network fails are improved.

The invention discloses a port operation vehicle scheduling method and system based on map data, and particularly relates to the technical field of scheduling optimization. Real-time map state information of a port channel network is generated by collecting real-time map data of the port channel network and analyzing the real-time occupation state of a port channel road section. And analyzing the local change trend of the occupation state of the port channel road section and the priority change of the vehicle task based on the real-time map state information, outputting dynamically updated vehicle task priority information and local area map data updating information, identifying a priority propagation abnormal area, and generating priority inversion risk early warning information. By identifying the vehicle path conflict points causing the traffic jam, generating the path conflict resolution strategy information, adjusting the vehicle path planning and executing the vehicle scheduling, the collaboration, responsiveness and traffic efficiency of the operation vehicle scheduling in the port environment are effectively improved.

The invention relates to the technical field of intelligent traffic systems, and provides an attention traffic signal control method and system based on a physical prior constraint time sequence diagram in order to solve the problems that in the prior art, risks cannot be recognized in advance, and direct quantification and constraint are lacked for core risks. The attention traffic signal control method based on the physical prior constraint time sequence diagram comprises the steps that the physical storage capacity of a road section is calculated, and the remaining storage space of a downstream road section is obtained in combination with the current vehicle number, namely overflow safety buffer; the overflow safety buffer is converted into overflow risk cost, and a constrained Markov decision process model with the passing efficiency as the return and the overflow risk as the cost is constructed; and solving the constrained Markov decision process model by using a pre-trained time sequence diagram attention network to obtain an optimal signal control action for balancing the traffic efficiency and the overflow risk. According to the method, the overflow risk can be proactively identified and actively avoided, so that cascade congestion caused by queue overflow can be actively avoided.

The invention provides a mine laneway traffic scheduling method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a continuous vibration signal monitored by a distributed optical fiber sound wave sensing system in a mine laneway; performing filtering and signal identification on the continuous vibration signal to identify a vehicle driving signal; monitoring the triggering time of the calibration point in the mine laneway according to the vehicle driving signal, and determining the position and the driving direction of the vehicle; according to the vehicle driving signal, the position of the vehicle and the driving direction, determining whether vehicle early warning needs to be carried out in the mine laneway; and in response to the need of vehicle early warning, generating a vehicle scheduling and early warning instruction so as to perform vehicle scheduling. Therefore, by analyzing the vibration signals monitored by the distributed optical fiber sound wave sensing system in the mine laneway in real time, the position, the speed and the vehicle driving direction of the vehicle are accurately recognized, so that before potential traffic conflicts occur, dispatching and early warning instructions are sent to related vehicles, and the passing efficiency and the safety of the mine laneway are effectively improved.

A cooperative merging method for a mixed traffic confluence area based on enhanced graph reinforcement learning comprises the following steps: firstly, designing a vehicle cooperation intensity model for effectively modeling the intensity of cooperative interaction between vehicles; then, based on a vehicle cooperation intensity model, combining a perception-communication relationship between vehicles with a vehicle-road cooperation communication relationship, constructing a local-global graph, depicting a dynamic cooperation relationship of a vehicle network from macroscopic and microscopic perspectives, and deeply mining driving behavior characteristics of the vehicles; then, a graph convolutional network combined with a Kolmogorov-Arnod network is adopted, and driving features of the vehicles are extracted and aggregated from the local-global graph; and finally, introducing a graph mutual information maximization method to optimize an iteration process of a graph reinforcement learning strategy, thereby ensuring that an accurate lane changing decision is generated for the CAVs. According to the method, the vehicle merging success rate and efficiency of the ramp merging area are improved, the possibility of congestion and collision of the ramp merging area is reduced, and the passing efficiency of the merging area is improved.

The invention provides a method for determining a motorcade driving strategy and electronic equipment, and relates to the technical field of path planning, and the method comprises the steps: firstly, determining a comprehensive operation index of a motorcade based on cooperation data and a basic operation index of the motorcade; secondly, the collaborative data and interaction data and prior data of the motorcade are fused to obtain fused data, the interaction data are data of the current running state of the vehicle and the road condition of the environment where the vehicle is located, and the prior data are used for representing historical driving data matched with the current scene; and finally, based on the fusion data, the comprehensive operation index and the path planning data of the motorcade, determining a target driving strategy of the motorcade, the path planning data being a constraint for supporting path planning. The technical problem that in the prior art, cooperative requirements of distance control, path conflict avoidance and the like of the motorcade in an off-road environment are not considered, so that the overall passing efficiency and safety of the motorcade are insufficient is solved.

The invention relates to the technical field of unmanned aerial vehicle flight control and intelligent navigation, in particular to an unmanned aerial vehicle low-altitude obstacle avoidance route planning method and system, and the method comprises the steps: firstly fusing three-dimensional space data and airspace rules, constructing a signed distance field, a capacity field and a state covariance, generating a channel graph, and calculating a door time lattice capacity upper limit; establishing joint trajectory optimization in the channel graph, combining geometric feasibility, dynamics reachable and covariance-based opportunity constraint separation, outputting a nominal trajectory, and controlling input and door crossing time; aligning the door penetrating time with the door time grid flow by a unification method to obtain the door time slot flow and the dual price; airborne control is executed according to a nominal trajectory and a time sequence constraint closed loop, and capacity change and price change are re-optimized and fed back in a time window under a triggering condition. According to the invention, the safety interval, the punctuality rate and the traffic efficiency are improved.

The invention provides a motorcade active-disturbance-rejection control method and system with a variable time interval strategy. The method comprises the following steps: acquiring state information of each vehicle in a vehicle queue system; the leader vehicle and the following vehicle carry out information transmission; under the condition of considering external disturbance, a vehicle dynamics model referring to the Newton second law is established, and a variable time interval strategy is constructed; compensating unmodeled dynamic and external disturbances in the vehicle dynamics model by using a time-varying gain expansion state observer in an active disturbance rejection control technology, and constructing a virtual controller and an actual controller based on a backstepping method; and according to the constructed virtual controller and the actual controller, realizing position and speed synchronization of the following vehicle and the leading vehicle in the queue under the condition of considering unmodeled dynamics and external disturbance. According to the method, the non-smooth problem at the saturation threshold is solved, the fluctuation influence can be effectively reduced, the safety is guaranteed, and the passing efficiency is improved; and the problem of gain attenuation is solved while the high gain characteristic is reserved.

The invention discloses an expressway construction area step-by-step variable speed limit control method based on a dual depth Q network, and the method comprises the steps: constructing an environment model interacting with a reinforcement learning agent, and the environment model comprises a state space, an action space and a reward function; constructing a reinforcement learning model based on a dual depth Q network, and generating a training data set through iterative interaction between the intelligent agent and the environment model; and training the dual depth Q network by using the training data set and taking maximization of accumulated rewards as a target, and realizing step-by-step variable speed limit control of the highway construction area based on a trained model. According to the method, data driving is taken as a main part, the overall income brought by step-by-step variable speed limit control is represented by comprehensively using driving safety, high efficiency and system stability indexes, a more intelligent and dynamic speed limit control strategy is constructed, and the traffic efficiency and the traffic safety level are improved.

The invention discloses a learning and model hybrid driven autonomous vehicle safety control method and system, which are suitable for expressway convergence control in a hybrid traffic environment. According to the method, multi-agent reinforcement learning and a safety potential field theory are combined to construct a hybrid control framework; collaborative decision and adaptive control of ramp and main line automatic driving vehicles are realized through a multi-agent near-end strategy optimization algorithm; meanwhile, a novel relatively safe potential field energy index is introduced, the interactive collision risk between the vehicles is quantified in real time, and a reward function is designed based on the index to guide an intelligent agent to learn safer behaviors; in addition, a dynamic safety intervention mechanism is further arranged, when risk assessment exceeds a preset threshold value, protective action coverage is automatically triggered, and risk avoidance is implemented. According to the method, the collision rate is remarkably reduced, meanwhile, high passing efficiency and reward revenue are kept, and the safety, coordination and deployability of the automatic driving vehicle are effectively improved.

The invention is suitable for the technical field of traffic management, and provides an urban traffic dynamic command method and system based on digital twinning. According to the method, the high-quality data set is generated through direct multi-source data acquisition, restoration and integration; hierarchical dynamic storage is carried out, and block chain security management is carried out; carrying out stored multi-source data acquisition and processing, and constructing a real-time dynamic traffic twin model; based on the real-time dynamic traffic twinborn model, decision optimization of traffic control is carried out, evaluation feedback is carried out through edge calculation, and model parameters of the real-time dynamic traffic twinborn model are corrected; a three-level data storage system is adopted, a data management center is constructed, and data life cycle management is carried out. According to the method, the real-time dynamic traffic twin model can be constructed by fusing multi-source data, traffic flow simulation, congestion prediction and dynamic optimization are realized, a scientific basis is provided for smart city traffic management, traffic congestion is effectively relieved, and the urban traffic efficiency is improved.

The invention relates to a motion planning method considering visibility cost in an urban multi-blind-area scene. According to the method, firstly, all blind areas in a scene are extracted, a blind area boundary set is generated in combination with road topology, and then key blind areas are screened out according to the driving intention of a vehicle. Then initializing, growing, updating and predicting a state subset of phantom traffic participants in the key blind area, then generating candidate paths, calculating the visibility cost of each path based on the state subset, combining other costs to obtain the total cost, and selecting the minimum cost as the optimal path. And finally, taking the optimal path as a reference, searching discrete acceleration actions to generate a speed curve meeting constraints, and jointly forming a motion plan of the self-vehicle by the two speed curves. Compared with the prior art, the method has the advantages of safety, passing efficiency, comfort and the like.

The invention provides a bus signal priority control method considering coordination of a phase deviation state, and belongs to the technical field of vehicle-road cooperation. The method comprises the following steps: constructing an end-side-cloud architecture system for an intersection implementing bus signal priority control, acquiring basic data of the intersection and buses, establishing a signal priority control method considering coordination phase deviation state constraint, and calculating the maximum bus signal priority time of the intersection. And signal timing schemes under different priorities are generated in advance according to the priority time, so that the accuracy and the real-time performance of bus priority execution are ensured. According to the invention, a reliable control means for passing without stopping is provided for buses, interference to social vehicle passing benefits is reduced, the coordination offset time is fully utilized, the success rate of bus signal priority control is increased, the riding experience of passengers can be improved, the energy consumption of the buses can be reduced, and the cost is reduced. The method is of great significance to further development of an urban public transport system.

The invention provides an intelligent traffic scheduling method and system based on multi-modal fusion and space-time prediction, and relates to the technical field of intelligent traffic scheduling, and the method comprises the steps: collecting traffic-related multi-modal data; performing preprocessing and data-level and feature-level fusion processing on the multi-modal data to generate a fusion feature matrix; constructing an STGCN-LSTM space-time prediction model according to the fusion feature matrix, and predicting the traffic flow and queuing length of each entrance lane of the current scene to obtain a prediction result; and dynamically optimizing a signal lamp timing scheme according to the prediction result and the real-time traffic state. By accurately sensing a real-time traffic state, predicting a traffic flow change trend in a short time in the future and dynamically adjusting an intersection signal lamp timing scheme, on-demand timing is realized, the average waiting time of vehicles is shortened, the queuing length at an intersection is reduced, the traffic efficiency at the intersection is improved, and meanwhile, the traffic demands of pedestrians and non-motor vehicles are considered; and the safety and smoothness of traffic operation are ensured.