AI-based adaptive traffic signal control to optimize mobility in cities

Abstract

An AI-based adaptive traffic signal control system (100) for optimizing urban mobility, comprising: (a) a plurality of traffic sensors and cameras configured to collect real-time traffic data at one or more intersections; (b) a preprocessing and data management module to clean and normalize the collected data; (c) a traffic pattern recognition module configured to classify traffic conditions based on real-time and historical traffic data; (d) an AI-based signal timing optimization module that uses reinforcement learning to dynamically determine optimal signal phase sequences and green time assignments; (e) a communication and control execution module configured to transmit and implement the optimized signal timing plans to traffic signal control devices; and (f) a monitoring and system feedback module configured to evaluate traffic performance and provide feedback to the AI-based signal timing optimization module, whereby the system dynamically adjusts traffic signal phase schedules in real time to improve traffic flow and reduce congestion.

Description

[0001] The present invention relates to the field of intelligent transportation systems, in particular to the adaptive control of traffic signals using artificial intelligence and real-time data analysis. It specifically relates to urban traffic management systems aimed at optimizing mobility, reducing congestion, and improving the efficiency of the road network. In particular, the invention relates to an AI-based adaptive traffic signal control system that dynamically adjusts signal timing based on real-time traffic conditions.

[0002] In urban areas around the world, severe traffic congestion is becoming increasingly common due to rapid population growth, increasing vehicle numbers, and expanding road networks. Conventional traffic signal control systems generally operate with predefined fixed schedules or limited daily adjustments that cannot account for real-time fluctuations in traffic volume. As a result, vehicles often have to wait at intersections longer than necessary, leading to increased fuel consumption, delays, and an increase in air pollution.

[0003] Recent advances in sensor technologies and data collection methods have enabled the availability of real-time traffic data. However, existing signal control systems are unable to efficiently utilize this data to dynamically adjust signal timing to current traffic conditions. Furthermore, most systems rely on rule-based algorithms or manual configurations that are not scalable and often require significant human intervention. Therefore, there remains a need for systems with improved responsiveness and less reliance on manual configurations.

[0004] Artificial intelligence techniques, particularly machine learning and reinforcement learning, have demonstrated significant potential for learning optimal control strategies from traffic flow patterns in real time. Integrating artificial intelligence into networked traffic signal infrastructures makes it possible to develop adaptive signal control systems that continuously learn from changing traffic environments and optimize phase timing in real time. Such intelligent systems address the shortcomings of existing time-dependent and active control systems by improving traffic mobility, reducing delays, and making better use of available infrastructure.

[0005] An object of the present disclosure is to enable real-time optimization of traffic signal timing based on actual traffic conditions.

[0006] Another objective of this disclosure is to reduce vehicle waiting time and overall delay at intersections.

[0007] Another objective of the present disclosure is to minimize fuel consumption and vehicle emissions through a smoother traffic flow.

[0008] Another object of the present disclosure is to provide a self-learning system that continuously adapts to changing traffic patterns without manual intervention.

[0009] Another objective of this disclosure is the efficient use of existing road infrastructure without the need for major structural changes.

[0010] Another objective of this disclosure is to ensure robust traffic performance during peak hours and in abnormal traffic scenarios.

[0011] Another objective of this disclosure is to provide a centralized interface for real-time monitoring and control by transport operators.

[0012] Another objective of this disclosure is to provide historical analyses and reports to support data-driven decision-making for urban planning.

[0013] Further objects and advantages of the present disclosure will become apparent from the following description, which is not intended to limit the scope of the present disclosure.

[0014] The present invention generally relates to an AI-based adaptive traffic signal control system configured to optimize vehicle mobility in urban road networks using real-time data and intelligent decision making.

[0015] An embodiment of the present invention provides a preprocessing and data management module that cleans and normalizes real-time traffic data to generate accurate signal control inputs.

[0016] Another embodiment of the invention includes a traffic pattern recognition module that analyzes historical and real-time traffic data to classify current traffic conditions and detect traffic spikes or abnormal events.

[0017] Another embodiment of the invention is an AI-based signal timing optimization module that uses reinforcement learning to dynamically select optimal signal phase sequences and green time assignments.

[0018] Another embodiment of the invention is a communication and control execution module configured to securely and in real time communicate and implement the optimized signal plans to roadside traffic controllers.

[0019] Another embodiment of the invention is a monitoring and feedback module coupled with a user interface to continuously evaluate traffic performance, provide closed-loop feedback to the AI ​​engine, and enable operators to view real-time analytics and manually override control actions when needed.

[0020] The present invention relates to an AI-controlled adaptive traffic signal control system (100) for optimizing urban traffic flow in real time. It comprises several integrated modules, including data acquisition, preprocessing, traffic pattern recognition, AI-based signal optimization, communication and execution, monitoring and feedback, and a user interface. Together, these modules collect and analyze real-time traffic data, intelligently generate optimal signal timing, implement it at intersections, and continuously refine the control strategy based on live feedback. Data acquisition and sensor module:

[0021] This module comprises a network of traffic sensors, cameras, and data interfaces for connected vehicles strategically installed at intersections and road sections. These sensor units continuously collect real-time information such as vehicle counts, queue lengths, average travel speed, and traffic density. The module ensures the seamless integration of heterogeneous data sources and transmits the aggregated traffic information to a central processing unit via a secure communication channel. Preprocessing and data management module:

[0022] Raw traffic data from various sources often contains noise, duplication, and missing values. This module performs data cleaning, filtering, normalization, and spatiotemporal adjustment to transform the raw data into structured and usable formats. It also performs short-term traffic condition forecasts using statistical approaches to fill in minor data gaps and stores the processed data in a scalable database for later analysis. Traffic pattern recognition module:

[0023] This module uses machine learning algorithms to analyze the processed traffic data and detect recurring traffic patterns, such as peak-hour congestion, special events, and abnormal traffic conditions. By continuously learning from historical and real-time data, the module dynamically classifies traffic conditions and provides the control algorithm with a contextual understanding of current road conditions. AI-supported module for optimizing signal times:

[0024] The heart of the invention is the AI-based decision engine, which uses reinforcement learning to determine the optimal signal phase sequences and green time assignments for each intersection. The reinforcement learning agent interacts with the traffic environment and receives feedback in the form of reduced queue lengths and delays, allowing it to continuously improve its control strategy. The module outputs the optimal signal timings tailored to the prevailing traffic conditions. Communication and control execution module:

[0025] This module is responsible for the secure forwarding of optimized signal schedules from the AI ​​model to the actual traffic signal controllers in the field. It ensures real-time synchronization between the control center and the intersection hardware, executes the new phase commands without interrupting service, and confirms successful implementation with acknowledgement signals. Monitoring and system feedback module:

[0026] Once the optimized signal control is implemented, this module monitors traffic performance in real time to evaluate the effectiveness of the implemented strategy. Key performance indicators such as average waiting time, intersection throughput, and stop rates are tracked and fed back to the AI-based optimization module. This closed-loop mechanism ensures that the system continuously adapts to changing traffic conditions. User interface and reporting module:

[0027] The system also includes a user-friendly dashboard for traffic operators and urban planners. This module presents real-time traffic information, historical performance analyses, and comparative reports through visual graphs and alerts. It facilitates manual interventions when needed and allows authorized users to adjust policy parameters, configure intersection priorities, and generate customized reports for informed decision-making.

[0028] The invention is explained again below with reference to the figure. It shows: Fig. : an illustration of an AI-controlled adaptive traffic signal control system (100) for optimizing urban traffic flow in real time.

[0029] Fig.illustrates an AI-controlled adaptive traffic signal control system (100) designed to optimize urban traffic flow in real time. The operation of the AI-based adaptive traffic signal control system begins with the data acquisition and sensing module, which continuously collects real-time traffic information such as vehicle count, queue length, and traffic density from sensors and cameras installed at key intersections. The collected raw data is transmitted to the control center, where the preprocessing and data management module cleans and normalizes the data sets to ensure consistent and accurate input. The refined information is then fed into the traffic pattern recognition module, which analyzes historical and real-time patterns to classify current traffic conditions and detect events such as peak hours or unusual congestion.Based on the detected traffic condition, the AI-based control algorithm in the AI-based signal timing optimization module determines the optimal signal phase sequence and green time allocation using reinforcement learning techniques to achieve maximum traffic flow efficiency.

[0030] Once the optimal timing parameters are established, the Communication and Control Execution Module transmits these timing plans to the signal controllers at the field level, ensuring real-time implementation without disrupting ongoing traffic operations. The Monitoring and System Feedback Module continuously evaluates the impact of the deployed control parameters by tracking performance indicators such as average waiting time, queue length, and intersection throughput. The feedback loop enables the AI ​​system to adaptively refine its decision-making policy as traffic conditions change throughout the day. Finally, the User Interface and Reporting Module provides traffic operators with a live view of current performance, detailed analytics, and alerts, and allows authorized users to configure priority settings and manually override system operations when needed.

Claims

[1] An AI-based adaptive traffic signal control system (100) for optimizing urban mobility, comprising: (a) a large number of traffic sensors and cameras configured to collect real-time traffic data at one or more intersections; (b) a preprocessing and data management module for cleaning and normalizing the collected data; (c) a traffic pattern recognition module configured to classify traffic conditions based on real-time and historical traffic data; (d) an AI-supported signal timing optimization module that uses amplification learning to dynamically determine optimal signal phase sequences and green time allocations; (e) a communication and control execution module configured to transmit and implement the optimized signal timings to traffic signal controllers; and (f) a monitoring and system feedback module configured to evaluate traffic performance and provide feedback to the AI-based signal timing optimization module, whereby the system dynamically adjusts traffic signal phase schedules in real time to improve traffic flow and reduce congestion. [2] System (100) according to claim 1, wherein the traffic sensors are selected from induction loop detectors, radar sensors, infrared sensors, Bluetooth sensors and connected vehicle interfaces. [3] System (100) according to claim 1, wherein the traffic pattern recognition module uses machine learning algorithms selected from k-means clustering, support vector machines and neural networks. [4] System (100) according to claim 1, wherein the reinforcement learning algorithm is selected from Q-Learning, Deep Q-Network (DQN) and Proximal Policy Optimization (PPO). [5] System (100) according to claim 1, wherein the AI-based signal timing optimization module is configured to minimize vehicle queue length and average intersection delay. [6] System (100) according to claim 1, wherein the communication and control execution module includes a secure wireless communication protocol for transmitting optimized signal plans to traffic signal control units. [7] System (100) according to claim 1, wherein the monitoring and system feedback module evaluates key performance indicators selected from average waiting time, stopover rate, queue length and intersection throughput. [8] System (100) according to claim 1 further comprises a user interface and reporting module configured to display traffic performance in real time and historical analyses to traffic operators. [9] System (100) according to claim 1, wherein the preprocessing and data management module performs temporal interpolation to estimate missing traffic data. [10] System (100) according to claim 1, wherein the traffic pattern recognition module continuously updates its classification model based on newly acquired traffic data.

Images