Intersection traffic sensing method and system based on holographic sensing intelligent edge calculation
By adopting holographic perception and intelligent edge computing in the intersection traffic perception system, combining multiple sensor data and intelligent decision-making modules, the shortcomings of intersection traffic perception and optimization in the existing technology are solved, and more efficient and flexible traffic management is achieved.
Patent Information
- Application Number
- CN202510055051.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-14
- Publication Date
- 2025-04-18
AI Technical Summary
It is difficult for the existing technology to effectively perform intersection traffic perception and optimization, especially when facing sudden traffic conditions, the flexibility and resilience of traffic management are insufficient.
The intersection traffic perception system based on holographic perception and intelligent edge computing is adopted, and the data is fused through multiple sensors (camera, lidar, sound sensor), combined with edge computing and intelligent decision-making modules, evaluate and adjust signal light strategies in real time.
It realizes more comprehensive and accurate traffic condition monitoring, improves the flexibility and resilience of traffic management, reduces traffic congestion and delays, and improves traffic efficiency.
Smart Images
Figure CN119851489A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of traffic intersection perception, and in particular to a method and system for traffic intersection perception based on holographic perception intelligent edge computing. Background Art
[0002] With the acceleration of urbanization and the surge in motor vehicles, the transportation system is facing unprecedented challenges. Traffic congestion has become a "chronic disease" in many large cities, which not only seriously affects the daily travel efficiency of citizens, increases the time and cost of commuting, but also aggravates air pollution and energy consumption. Frequent traffic accidents are also a major pain point, bringing huge economic burdens and psychological trauma to families and society.
[0003] The development of big data and intelligent systems provides people with the ability to provide early traffic warnings. However, faced with congestion that may occur at any time, how to perceive and optimize traffic at intersections remains a key challenge. Summary of the invention
[0004] In view of the deficiencies in the prior art, the present invention provides a traffic perception method and system for intersections based on holographic perception intelligent edge computing, which provides a more comprehensive and accurate traffic condition through the fusion of multiple sensors (such as cameras, lidars, sound sensors, etc.); this multi-dimensional perception capability can effectively capture complex traffic environments and improve monitoring accuracy. It also relies on intelligent decision-making modules combined with holographic perception data to evaluate and adjust traffic light strategies in real time to respond to sudden traffic conditions (such as accidents or special events), thereby enhancing the flexibility and adaptability of traffic management.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions: The intersection traffic perception system based on holographic perception and edge computing includes: A variety of sensor modules to collect intersection traffic data, including cameras, lidar, and sound sensors; Edge computing modules, deployed near intersections, are used to process and analyze data collected by sensors in real time; Data fusion module, which integrates feature data from different sensors to generate comprehensive traffic status; Intelligent decision-making module optimizes traffic light timing strategy based on fused data and reinforcement learning algorithm.
[0006] Preferably, the edge computing module is configured with a real-time data processing algorithm to reduce the response delay of traffic light adjustment.
[0007] Preferably, the intelligent decision-making module adopts Learning algorithm, dynamically updated according to traffic status values to optimize the traffic light strategy.
[0008] The intersection traffic perception method based on holographic perception and edge computing includes the following steps: Collect real-time intersection traffic data through multiple sensors, including vehicle counts, speeds, delays, and noise levels; Use edge computing to pre-process, denoise and extract features from the collected data to generate key traffic features; integrating the traffic characteristic data from different sensors through a data fusion algorithm to generate a comprehensive traffic state estimate; Based on the traffic state estimation, a reinforcement learning algorithm is used to dynamically adjust the traffic light timing strategy to optimize traffic flow and reduce delays.
[0009] Preferably, the data fusion algorithm adopts Kalman filtering technology, and the state estimation formula is:
[0010] in, is the current state estimate, is the Kalman gain, for The measured value at the moment; H is the observation matrix; For Momentary status prediction The state estimate at the moment.
[0011] Preferably, the reinforcement learning algorithm adopts Learning algorithm, the update formula is:
[0012] in: Indicates immediate reward; value Indicates in status Take action After that, the estimated cumulative rewards that can be obtained; is the learning rate, which ranges from 0 to 1; is the discount factor, and the value range of the discount factor is between 0 and 1; is the new state after taking the action; It is a new state All possible actions of The maximum value of the value.
[0013] Preferably, the instant reward It is calculated by the following formula:
[0014] in, is the vehicle flow, indicating the time The number of vehicles passing through the intersection; Delay cost: represents the delay and driving time loss caused by the change of traffic lights.
[0015] Preferably, the vehicle flow The calculation formula is:
[0016] in, For in time Neidi The number of vehicles passing in each direction; Delay costs The calculation formula is:
[0017] in, For in time Neidi The average delay in each direction.
[0018] Preferably, the method comprises using a sound sensor to monitor traffic noise to determine changes in traffic flow, and the determination process comprises the following steps: Collecting noise signals from sound sensors; Analyze the noise frequency distribution by fast Fourier transform; Based on the preset noise threshold, the traffic flow changes are judged and corresponding traffic status feedback is generated.
[0019] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a method and system for intersection traffic perception based on holographic perception intelligent edge computing, which provides more comprehensive and accurate traffic conditions through the fusion of multiple sensors (such as cameras, lidar, sound sensors, etc.); this multi-dimensional perception capability can effectively capture complex traffic environments and improve monitoring accuracy.
[0020] The edge computing module of the present invention processes and analyzes data at a location closer to the data source, reducing delays, allowing the adjustment of traffic signals to respond to current traffic conditions in real time, quickly optimizing signal timing, and improving traffic efficiency.
[0021] The intelligent decision-making module of the present invention, combined with holographic perception data, can evaluate and adjust traffic light strategies in real time to respond to sudden traffic situations (such as accidents or special events), thereby enhancing the flexibility and adaptability of traffic management.
[0022] The present invention collects data through multiple sensors, provides redundant information, and reduces the risk of system failure caused by a single sensor failure. Edge computing can perform redundant data analysis locally to ensure the stability and security of decision-making.
[0023] By monitoring traffic flow and delays in real time, the system can effectively allocate signal time, improve resource utilization in different directions, reduce unnecessary waiting time, and reduce energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0025] Figure 1 It is a schematic diagram of the system flow of the present invention. DETAILED DESCRIPTION
[0026] The present invention is further described in detail below with reference to the accompanying drawings.
[0027] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be used for other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.
[0028] See also Figure 1 ,The intersection traffic perception system based on holographic perception and edge computing includes, A variety of sensor modules to collect intersection traffic data, including cameras, lidar, and sound sensors; Edge computing modules, deployed near intersections, are used to process and analyze data collected by sensors in real time; Data fusion module, which integrates feature data from different sensors to generate comprehensive traffic status; Intelligent decision-making module optimizes traffic light timing strategy based on fused data and reinforcement learning algorithm.
[0029] The various sensor modules of the present invention are composed of cameras, laser radars, sound sensors, etc., and are used for data collection. Specifically:
[0030] Use cameras to capture traffic flow and vehicle types and generate video stream data; Use LiDAR to measure distance, obtain vehicle position and speed, and generate LiDAR point cloud data; Use sound sensors to monitor traffic noise and generate sound signal data to determine traffic conditions.
[0031] The data fusion module of the present invention integrates the characteristic data from different sensors to generate a comprehensive traffic status Among them, the intelligent decision-making module of the present invention adopts Learning algorithms (i.e. ), dynamically updated according to traffic conditions values to optimize the traffic light strategy.
[0032] The present invention also includes a feedback and optimization module: including data aggregation and uploading as well as feedback and optimization; Data aggregation and upload The important data that has been processed and analyzed is summarized and sent to the cloud for long-term storage and further analysis. The edge computing module determines the upload frequency and data type to optimize network usage.
[0033] Feedback and Optimization: By collecting feedback and historical data from the user interface, edge devices continuously optimize their algorithms and decision-making rules to improve future response efficiency and accuracy.
[0034] Based on the above system, the present invention also discloses a method for sensing traffic at intersections based on holographic sensing and edge computing, and the method comprises the following steps: Step 1: Collect intersection traffic data in real time through multiple sensors, including vehicle counts, speeds, delays, and noise levels; Step 2: Use edge computing to pre-process, denoise and extract features from the collected data to generate key traffic features; Step 3: integrating the traffic characteristic data from different sensors through a data fusion algorithm to generate a comprehensive traffic state estimation; Step 4: Based on the traffic state estimation, a reinforcement learning algorithm is used to dynamically adjust the traffic light timing strategy to optimize traffic flow and reduce delays.
[0035] Furthermore, the method uses a sound sensor to monitor traffic noise to determine changes in traffic flow, and the determination process includes the following steps: Collecting noise signals from sound sensors; Analyze the noise frequency distribution by fast Fourier transform; Based on the preset noise threshold, the traffic flow changes are judged and corresponding traffic status feedback is generated.
[0036] In step 1 of the method, traffic data at the intersection, including vehicle count, speed, delay time and noise level, are collected in real time through multiple sensors, and a data set is established; In step 2 of this method, edge computing is used to preprocess, denoise and extract features from the collected data to generate key traffic features. The detailed steps are as follows: (1) Data preprocessing: Data cleaning, using the outlier detection method Z-score to remove abnormal data, and setting a reasonable Z-score threshold based on the characteristics and requirements of the data set. For outliers that exceed the threshold, delete them; (2) Data denoising: denoising data that is erroneous or deviates from expected values, using regression methods to smooth data and reduce the interference of noise on model training; (3) Data normalization unit, which is used to normalize vehicle count, speed, delay time and noise level data, and uses Z-score for normalization calculation; (4) Feature extraction: The data format is converted through data normalization to eliminate differences in dimension and value range; the most representative features are selected from the normalized data to generate key traffic features.
[0037] In step 3 of this method, the data fusion algorithm uses the Kalman filter algorithm to integrate data from different sensors to generate accurate traffic state estimation. The state estimation formula is:
[0038] in, is the current state estimate, is the Kalman gain, for The measured value at the moment; H is the observation matrix; For Momentary status prediction The state estimate at the moment.
[0039] In step 4 of the present invention, traffic optimization includes using a reinforcement learning algorithm (Q-learning) to optimize the signal timing strategy. The algorithm continuously tests different signal timing strategies, evaluates their impact on traffic flow and delay, and improves the overall performance of the system by updating the strategy.
[0040] Reinforcement learning algorithm uses Learning algorithms, The value update formula is:
[0041] in: Indicates immediate reward; value Indicates in status Take action After that, the estimated cumulative rewards that can be obtained; is the learning rate, which ranges from 0 to 1. A smaller learning rate (such as 0.01 or 0.001) will make the learning process more robust, but converge more slowly, while a larger learning rate (such as 0.1 or 0.5) may speed up the learning process but may cause instability or divergence. is the discount factor, and the value range of the discount factor is between 0 and 1; is equal to 1, meaning that all future rewards are equally important (suitable for long-term goals), while Close to 0 means more attention to immediate rewards; is the new state after taking the action; It is a new state All possible actions of The maximum value of the value.
[0042] The decision on strategies to optimize traffic flow and reduce delay costs is calculated and analyzed through the following steps: 1. Calculate traffic flow
[0043] Traffic flow The expression in time The number of vehicles passing through the intersection, which is collected by sensors in real time; traffic flow The calculation formula is:
[0044] in, For in time Neidi The number of vehicles passing in each direction.
[0045] 2. Calculate the cost of delay
[0046] Delay Cost Delays and lost travel time due to signal changes; delay costs Calculated by:
[0047] in, For in time Neidi The average delay in each direction.
[0048] 3. Optimize the objective function The optimization goal is to maximize traffic flow , while minimizing the cost of delay Achieve, by defining a comprehensive indicator to achieve, immediate reward , Indicates the current status (including traffic flow, signal light status, etc.), Indicates possible decisions (including adjusting traffic light timing); immediate rewards It is calculated by the following formula:
[0049] when When it is a positive value, it means that the current decision improves traffic flow and reduces delays. The system will adjust the signal timing strategy based on this feedback.
[0050] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.
Claims
1. The intersection traffic perception system based on holographic perception and edge computing is characterized by: include, A variety of sensor modules to collect intersection traffic data, including cameras, lidar, and sound sensors; Edge computing modules, deployed near intersections, are used to process and analyze data collected by sensors in real time; Data fusion module, which integrates feature data from different sensors to generate comprehensive traffic status; Intelligent decision-making module optimizes traffic light timing strategy based on fused data and reinforcement learning algorithm.
2. The intersection traffic perception system based on holographic perception and edge computing according to claim 1 is characterized in that: The edge computing module is configured with a real-time data processing algorithm to reduce the response delay of traffic light adjustment.
3. The intersection traffic perception system based on holographic perception and edge computing according to claim 2 is characterized in that: The intelligent decision-making module adopts Learning algorithm, dynamically updated according to traffic status values to optimize the traffic light strategy.
4. A method for sensing traffic at intersections based on holographic sensing and edge computing using any of the systems described in claims 1-3, characterized in that: The following steps are involved: Collect real-time intersection traffic data through multiple sensors, including vehicle counts, speeds, delays, and noise levels; Use edge computing to pre-process, denoise and extract features from the collected data to generate key traffic features; integrating the traffic characteristic data from different sensors through a data fusion algorithm to generate a comprehensive traffic state estimate; Based on the traffic state estimation, a reinforcement learning algorithm is used to dynamically adjust the traffic light timing strategy to optimize traffic flow and reduce delays.
5. The method for sensing traffic at intersections based on holographic sensing and edge computing according to claim 4 is characterized in that: The data fusion algorithm adopts Kalman filtering technology, and the state estimation formula is: ； in, is the current state estimate, is the Kalman gain, for The measured value at the moment; H is the observation matrix; For Momentary status prediction The state estimate at the moment.
6. The method for sensing traffic at intersections based on holographic sensing and edge computing according to claim 5 is characterized in that: The reinforcement learning algorithm adopts Learning algorithm, the update formula is: ； in: Indicates immediate reward; value Indicates in status Take action After that, the estimated cumulative rewards that can be obtained; is the learning rate, which ranges from 0 to 1; is the discount factor, and the value range of the discount factor is between 0 and 1; is the new state after taking the action; It is a new state All possible actions of The maximum value of the value.
7. The method for sensing traffic at intersections based on holographic sensing and edge computing according to claim 6, characterized in that: The instant reward It is calculated by the following formula: ； in, is the vehicle flow, indicating the time The number of vehicles passing through the intersection; Delay cost: represents the delay and driving time loss caused by the change of traffic lights.
8. The method for sensing traffic at intersections based on holographic sensing and edge computing according to claim 6, characterized in that: The vehicle flow The calculation formula is: ； in, For in time Neidi The number of vehicles passing in each direction; Delay costs The calculation formula is: ； in, For in time Neidi The average delay in each direction.
9. The method for sensing traffic at intersections based on holographic sensing and edge computing according to claim 4, characterized in that: The method further includes using a sound sensor to monitor traffic noise to determine changes in traffic flow, wherein the determination process includes the following steps: Collecting noise signals from sound sensors; Analyze the noise frequency distribution by fast Fourier transform; Based on the preset noise threshold, the traffic flow changes are judged and corresponding traffic status feedback is generated.