DESC:SYSTEM FOR ANALYZING AND CLASSIFYING DRIVING PATTERNS OF A TWO-WHEELER
CROSS REFERENCE TO RELATED APPLICTIONS
The present application claims priority from Indian Provisional Patent Application No. 202421020618 filed on 19-03-2024, the entirety of which is incorporated herein by a reference.
TECHNICAL FIELD
The present disclosure generally relates to vehicular data analysis. Further, the present disclosure particularly relates to a system for analyzing and classifying the driving patterns of a two-wheeler.
BACKGROUND
The field of vehicular data analysis has evolved to address the growing need for assessing driving behavior, traffic management, and accident prevention. Two-wheelers form a significant portion of road traffic, necessitating methods to monitor and classify riding patterns. Various techniques have been explored for analyzing vehicular movement using sensor-based data acquisition, machine learning techniques, and cloud-based analytics.
One known technique employs inertial measurement units (IMUs) to capture motion-related parameters, such as acceleration and angular velocity. Such systems rely on predefined threshold-based models to classify riding behavior. However, such techniques exhibit limitations in handling variations caused by road conditions, sensor drift, and environmental factors. Additionally, such methods lack the capability to incorporate external influences, such as real-time traffic congestion and road conditions, which play a significant role in defining driving patterns.
Another approach integrates global positioning system (GPS) units along with speed and acceleration sensors. Such an approach records vehicular movement and applies rule-based or clustering techniques to classify riding patterns. Although such a method offers broader spatial analysis, challenges arise in handling dynamic variations in road conditions. Furthermore, the absence of visual data limits the ability to differentiate between genuine road-induced variations and deliberate rider actions.
Machine learning-based classification techniques have been explored to enhance driving behavior assessment. Supervised learning models have been employed to categorize riding styles based on historical datasets. While such approaches improve classification accuracy, reliance on a single sensor data source restricts adaptability across different environments. Further, existing techniques lack efficient data transmission mechanisms and cloud-based storage for scalable real-time analysis.
Video-based analysis has also been investigated, wherein onboard cameras record road conditions and assess rider behavior. Video processing techniques analyze lane discipline, traffic rule violations, and sudden manoeuvres. However, such methods require high computational resources and extensive storage capacity. Moreover, integration with additional sensor data remains limited, reducing overall effectiveness.
Other known techniques integrate multiple sensors and data analysis methodologies. However, such techniques present challenges, including increased hardware complexity, computational inefficiencies, and data synchronization issues. Current approaches lack an optimal combination of multi-sensor data acquisition, machine learning-based classification, and cloud-based storage for an accurate and scalable solution.
In light of the above discussion, there exists an urgent need for solutions that overcome the problems associated with conventional systems and/or techniques for analyzing and classifying driving behavior.

SUMMARY
The aim of the present disclosure is to provide a system for analyzing and classifying the driving patterns of a two-wheeler to enhance road safety, assess driver behavior, optimize traffic management, facilitate insurance risk assessment, improve fuel efficiency, and predict hazards.
The present disclosure relates to a system for analyzing and classifying the driving patterns of a two-wheeler. A plurality of sensors, including an accelerometer, a gyroscope, a global positioning system (GPS) unit, and a camera, gathers data related to speed, acceleration, orientation, location, and visual driving conditions. A communication interface transmits the gathered data to a server. The server receives the transmitted data, applies a clustering technique to categorize the received data into multiple clusters, and executes at least one supervised machine learning model selected from a group consisting of Random Forest classifier, XGBoost, and K-nearest neighbor to analyze each cluster and classify the driving patterns into categories, wherein such categories comprise aggressive behaviors and non-aggressive behaviors. The server provides the classified driving patterns to one or more external display devices. A cloud storage unit, operatively connected to the server, stores the classified driving patterns, a traffic congestion, and the road conditions.
Further, the server extracts multiple contextual factors, including a time of day, a weather condition, and a traffic density, using data obtained from an external application programming interface (API) and a plurality of sensors. The server correlates the extracted contextual factors with the classified driving patterns to identify specific behavioral characteristics of a driver. Correlating the contextual factors with the classified driving patterns enables a behavioral assessment.
Moreover, the server identifies specific stress-inducing conditions by analyzing abrupt deviations in the driving patterns correlated with contextual data. The server provides stress-reducing feedback through an external display device. Identifying stress-inducing conditions enables a real-time intervention to improve driver awareness and response.
Furthermore, the server identifies patterns indicating a drowsiness status by analyzing prolonged steady driving behavior combined with contextual data. The server alerts the driver through an external display device. Detecting drowsiness patterns enables proactive safety measures to reduce risks.
Additionally, the server compares the classified driving patterns with regional driving norms stored in a cloud storage unit to identify deviations and provide a region-specific driving behavior assessment. Comparing the classified driving patterns with regional norms enables the identification of location-based driving tendencies and regulatory compliance.
Further, the server analyzes transitions between classified driving patterns to identify potential triggers for aggressive behavior. The server suggests mitigation strategies through an external display device. Identifying the triggers for aggressive behavior enables preventive strategies to reduce unsafe driving actions.
Moreover, the server predicts collision risks by analyzing the classified driving patterns, the contextual factors, and the proximity data gathered by a plurality of sensors. Predicting collision risks enables preemptive alerts to prevent accidents.
Furthermore, the server generates anonymized driving pattern reports for urban planning applications and traffic management applications based on aggregated data stored in a cloud storage unit. Generating anonymized reports enables data-driven insights for infrastructure development and policy formulation.
Additionally, the server analyzes correlations between aggressive driving patterns and a vehicle fuel consumption. The server provides eco-driving recommendations through an external display device. Analyzing the correlation between aggressive driving patterns and fuel consumption enables optimal driving behavior for fuel efficiency.
In another aspect, the present disclosure provides a method for analyzing and classifying driving patterns of a two-wheeler. Data related to speed, acceleration, orientation, location, and visual driving conditions is gathered using a plurality of sensors, wherein the sensors comprise an accelerometer, a gyroscope, a global positioning system (GPS) unit, and a camera. The gathered data is transmitted from a plurality of sensors to a server through a communication interface. The transmitted data is received at the server. A clustering technique is applied to categorize the received data into multiple clusters. At least one supervised machine learning model, selected from a group consisting of Random Forest classifier, XGBoost, and K-nearest neighbor, is executed to analyze each cluster and classify the driving patterns into categories, wherein such categories comprise aggressive behaviors and non-aggressive behaviors. The classified driving patterns, a traffic congestion, and the road conditions are stored in a cloud storage unit operatively connected to the server. The classified driving patterns are provided to at least one external display device.
BRIEF DESCRIPTION OF DRAWINGS
The summary above, as well as the following detailed description of illustrative embodiments, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the present disclosure, exemplary constructions of the disclosure are shown in the drawings. However, the present disclosure is not limited to specific methods and instrumentalities disclosed herein. Moreover, those in the art will understand that the drawings are not to scale. Wherever possible, like elements have been indicated by identical numbers.
Embodiments of the present disclosure will now be described, by way of example only, with reference to the following diagrams wherein:
FIG. 1 illustrates a system for analyzing and classifying the driving patterns of a two-wheeler, in accordance with the embodiments of the present disclosure.
FIG. 2 illustrates a method for analyzing and classifying driving patterns of a two-wheeler, in accordance with the embodiments of the present disclosure.
FIG. 3 illustrates a sequence diagram of a system for analyzing and classifying the driving patterns of a two-wheeler, in accordance with the embodiments of the present disclosure.
In the accompanying drawings, an underlined number is employed to represent an item over which the underlined number is positioned or an item to which the underlined number is adjacent. A non-underlined number relates to an item identified by a line linking the non-underlined number to the item. When a number is non-underlined and accompanied by an associated arrow, the non-underlined number is used to identify a general item at which the arrow is pointing.
DETAILED DESCRIPTION
The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognise that other embodiments for carrying out or practising the present disclosure are also possible.
The description set forth below in connection with the appended drawings is intended as a description of certain embodiments of a system for analyzing and classifying the driving patterns of a two-wheeler and is not intended to represent the only forms that may be developed or utilised. The description sets forth the various structures and/or functions in connection with the illustrated embodiments; however, it is to be understood that the disclosed embodiments are merely exemplary of the disclosure that may be embodied in various and alternative forms. The figures are not necessarily to scale; some features may be exaggerated or minimised to show details of particular components. Therefore, specific structural and functional details disclosed herein are not to be interpreted as limiting, but merely as a representative basis for teaching one skilled in the art to variously employ the present invention.
While the disclosure is susceptible to various modifications and alternative forms, specific embodiment thereof has been shown by way of example in the drawings and will be described in detail below. It should be understood, however, that it is not intended to limit the disclosure to the particular forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternatives falling within the scope of the disclosure.
In the following detailed description of the embodiments of the disclosure, reference is made to the accompanying drawings, and which are shown by way of illustration specific embodiments in which the disclosure may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the disclosure, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present disclosure. The following description is, therefore, not to be taken in a limiting sense.
The present disclosure will be described herein below with reference to the accompanying drawings. In the following description, well known functions or constructions are not described in detail since they would obscure the description with unnecessary detail.
The term "system" refers to an arrangement of interconnected components to perform a specific function related to processing, analyzing, and classifying data. Such a system comprises multiple elements working collectively to achieve an intended operation. A system is implemented in hardware, software, or a combination thereof. The implementation of such a system depends on the required functionality, computational complexity, and deployment environment. In the context of data processing, a system consists of input components responsible for collecting data, processing components executing predefined logic, and output components delivering processed results.
The term "sensor" refers to a device that detects, measures, and responds to physical, chemical, or biological stimuli and converts such stimuli into electrical, optical, or mechanical signals for further processing. A sensor operates based on a specific sensing mechanism depending on the type of parameter being measured. Sensors are categorized into mechanical, thermal, optical, acoustic, and chemical sensors. A mechanical sensor detects force, pressure, strain, or displacement using resistive, capacitive, or piezoelectric sensing techniques. A thermal sensor measures temperature variations using a thermocouple, resistance temperature detector, or infrared sensor. An optical sensor detects light intensity, color, or wavelength variations using a photodiode, charge-coupled device, or spectrometer. An acoustic sensor captures sound waves or vibration patterns using a microphone or ultrasonic transducer. A chemical sensor identifies the presence or concentration of chemical substances using electrochemical, optical, or semiconductor-based detection techniques. A sensor operates independently or as part of a larger system, transmitting acquired data for further processing. The placement and integration of sensors depend on the application and environmental conditions. A gyroscope detects orientation, an accelerometer measures motion, and a GPS unit determines location coordinates.
The term "communication interface" refers to an electronic component responsible for establishing data exchange between different devices or subsystems. Such a communication interface supports wired or wireless communication technologies depending on the system requirements. A wired communication interface comprises serial, parallel, and network-based connections using transmission standards such as Universal Serial Bus, Ethernet, or Controller Area Network. A wireless communication interface comprises radio frequency-based technologies such as Bluetooth, Wi-Fi, and cellular networks. The selection of a communication interface depends on factors including data transmission rate, power consumption, distance coverage, and interoperability with other system components.
The term "server" refers to a computing device or system that processes, stores, and manages data received from multiple sources. A server executes predefined operations on received data and delivers the processed results to connected devices or users. Servers are classified based on function, including database servers, application servers, web servers, and computation servers. A database server stores data, executes queries, and provides requested information. An application server executes software applications and delivers services to client devices. A web server hosts and manages web-based content, handling user requests through Hypertext Transfer Protocol. A computation server performs high-performance computing tasks, processing large datasets and executing complex algorithms. A server comprises a central processing unit, memory, storage devices, and network interfaces to enable data communication with external devices. The implementation of a server is on-premises or cloud-based, depending on the required scalability and availability.
The term "cloud storage unit" refers to a digital storage infrastructure enabling the storage, retrieval, and management of data on remote servers accessible over a network. Such a cloud storage unit provides scalable and distributed storage architecture for managing large volumes of structured or unstructured data. The cloud storage unit is categorized into object storage, block storage, and file storage. Object storage organizes data as distinct objects with metadata and unique identifiers, commonly used for storing multimedia files, backups, and large datasets. Block storage segments data into fixed-size blocks, allowing fast retrieval and modification of stored information. File storage arranges data hierarchically in directories, supporting traditional file-based access. The cloud storage unit operates using networked storage clusters, enabling redundancy and fault tolerance. A distributed file system provides scalable storage access, a data lake enables large-scale analytics, and a content delivery network facilitates efficient data retrieval across geographically dispersed locations. The cloud storage unit interfaces with computing applications, enabling remote access and retrieval of stored data for further analysis. In a data processing system, a cloud storage unit stores classified data, event logs, and analytical results for future reference.
The term "clustering technique" refers to a data analysis method for categorizing a dataset into multiple groups based on similarities or patterns. Clustering techniques are categorized into hierarchical clustering, partitioning clustering, and density-based clustering. Hierarchical clustering constructs a nested hierarchy of clusters using agglomerative or divisive approaches. Partitioning clustering divides a dataset into a predefined number of clusters based on optimization criteria such as centroid minimization. Density-based clustering identifies regions of high data density, differentiating clustered data points from noise.
The term "machine learning model" refers to a computational method that enables automated pattern recognition and predictive analysis based on training data. A machine learning model operates by learning from historical data, identifying relationships between input features and outcomes, and making predictions for new data samples. Machine learning models are categorized into supervised learning, unsupervised learning, and reinforcement learning models. A supervised learning model, including decision trees, support vector machines, and artificial neural networks, requires labeled training data to develop predictive functions. An unsupervised learning model, including clustering algorithms and dimensionality reduction techniques, identifies hidden patterns in unlabeled data. A reinforcement learning model optimizes decision-making strategies through iterative interactions with an environment.
The term "external display device" refers to an output unit that visually presents processed data for user interaction or analysis. An external display device is implemented using display technologies such as liquid crystal displays, organic light-emitting diode panels, and projection-based screens. The selection of an external display device depends on resolution, refresh rate, power consumption, and interaction capabilities. A digital dashboard presents real-time telemetry data in automotive applications. A smart monitor provides visual analytics in data visualization systems. An augmented reality interface overlays digital information onto a field of view of user in interactive applications. In a data analysis system, an external display device presents classified driving patterns, alerts, and recommendations.
FIG. 1 illustrates a system 100 for analyzing and classifying the driving patterns of a two-wheeler, in accordance with the embodiments of the present disclosure. The system 100 comprises a plurality of sensors 102 that gather data related to speed, acceleration, orientation, location, and visual driving conditions of a two-wheeler. The plurality of sensors 102 comprises an accelerometer 104, a gyroscope 106, a global positioning system (GPS) unit 108, and a camera 110. The accelerometer 104 detects linear acceleration of the two-wheeler in multiple axes, allowing the identification of changes in speed and motion patterns. The gyroscope 106 measures angular velocity, enabling the determination of rotational movements and lean angles of the two-wheeler. The GPS unit 108 provides real-time location data, facilitating the tracking of routes, positioning accuracy, and speed estimation. The camera 110 captures visual data corresponding to the surrounding environment, including road conditions, obstacles, traffic signals, and lane markings. The plurality of sensors 102 operates individually or in combination to provide a dataset representing the movement characteristics of the two-wheeler. The collected data undergoes processing for further classification and analysis, enabling the identification of different driving behaviors and contextual factors. The plurality of sensors 102 operates continuously or at predefined intervals based on system requirements, external conditions, or specific trigger events. The plurality of sensors 102 transmits acquired data to a processing unit for real-time or offline analysis. The placement of the plurality of sensors 102 on the two-wheeler is determined based on measurement accuracy, environmental exposure, and integration feasibility. The plurality of sensors 102 interfaces with other components of the system 100, providing essential data for driving pattern classification and behavioral analysis.
In an embodiment, the communication interface 112 enables transmission of data gathered by the plurality of sensors 102 to the server 114. The communication interface 112 supports wired or wireless data transfer, depending on the system architecture, transmission distance, and power constraints. The communication interface 112 operates using various data exchange methods, including serial communication, Bluetooth, Wi-Fi, or cellular networks, to establish a reliable connection between the plurality of sensors 102 and the server 114. The communication interface 112 facilitates continuous or batch-mode data transmission based on system requirements and network conditions. The communication interface 112 implements data integrity mechanisms such as error detection and correction, encryption, and compression to maintain the accuracy and security of transmitted information. The communication interface 112 manages communication bandwidth and prioritizes data transmission to enable efficient data flow without latency issues. The communication interface 112 enables bi-directional communication, allowing the server 114 to send control commands, configuration updates, or acknowledgment signals to the plurality of sensors 102. The communication interface 112 dynamically adjusts transmission parameters based on environmental factors such as signal interference, network congestion, and power availability. The communication interface 112 integrates with cloud-based or local computing environments, facilitating real-time monitoring and historical data storage. The communication interface 112 establishes a communication link with multiple external devices, allowing data sharing and interoperability with third-party applications, traffic management systems, or user interfaces. The communication interface 112 supports secure authentication and encryption techniques, preventing unauthorized access to transmitted data.
In an embodiment, the server 114 receives transmitted data from the plurality of sensors 102 and processes the acquired information for analyzing and classifying the driving patterns of the two-wheeler. The server 114 applies a clustering technique to categorize received data into multiple clusters based on predefined similarity metrics, behavioral traits, or statistical properties. The server 114 analyzes patterns in the sensor data to identify variations in acceleration, braking, cornering, lane changes, and other driving characteristics. The server 114 executes at least one supervised machine learning model selected from a group consisting of Random Forest classifier, XGBoost, and K-nearest neighbor to analyze each cluster and classify the driving patterns into categories. The server 114 processes large volumes of historical and real-time data, adapting classification models based on continuously updated datasets. The server 114 utilizes training datasets containing labeled driving behaviors to improve classification accuracy and decision-making reliability. The server 114 applies feature selection techniques to identify the most relevant sensor parameters contributing to driving pattern classification. The server 114 performs statistical validation to assess classification accuracy and detect inconsistencies in processed data. The server 114 identifies patterns associated with aggressive behaviors, including sudden acceleration, harsh braking, sharp turns, and erratic lane changes. The server 114 distinguishes non-aggressive behaviors, including smooth acceleration, controlled braking, stable cornering, and steady speed maintenance. The server 114 generates predictive insights related to driver behavior, road safety, and traffic conditions based on classified driving patterns. The server 114 adapts classification models dynamically, incorporating real-time feedback and contextual information to refine pattern recognition accuracy. The server 114 performs computational optimizations to enhance processing efficiency, reducing latency in data classification and behavioral analysis.
In an embodiment, the server 114 provides the classified driving patterns to one or more external display devices 116, enabling visualization of analyzed data and driving behavior insights. The external display devices 116 receive processed data from the server 114 and present classified driving patterns in graphical, tabular, or alert-based formats. The external display devices 116 comprise digital dashboards, mobile applications, infotainment screens, and heads-up displays. The external display devices 116 display categorized driving behaviors, risk assessments, and contextual alerts in real time. The external display devices 116 indicate aggressive or non-aggressive driving behaviors using color-coded indicators, warning messages, or statistical summaries. The external display devices 116 provide interactive feedback to the user, enabling adjustments in driving habits based on classification results. The external display devices 116 integrate with other vehicle systems to enhance situational awareness and driver assistance features. The external display devices 116 support customization options, allowing users to modify display preferences, alert thresholds, and visualization layouts. The external display devices 116 synchronize with cloud storage services, enabling historical data retrieval and trend analysis. The external display devices 116 incorporate touch-screen interfaces, voice commands, or gesture recognition for user interaction and control. The external display devices 116 facilitate communication between the driver and external systems, including traffic management centers, insurance platforms, and fleet monitoring applications. The external display devices 116 generate reports summarizing classified driving patterns for behavioral assessment, insurance risk evaluation, or regulatory compliance. The external display devices 116 transmit user feedback to the server 114, enabling continuous improvement of classification models and data interpretation methodologies.
In an embodiment, the cloud storage unit 118 stores the classified driving patterns, traffic congestion data, and road conditions for historical analysis and future reference. The cloud storage unit 118 maintains a structured database to store categorized driving behaviors, enabling retrieval and review of past driving events. The cloud storage unit 118 organizes stored data using indexing and metadata tagging to facilitate efficient search and retrieval operations. The cloud storage unit 118 implements redundancy and backup mechanisms to prevent data loss due to system failures or network disruptions. The cloud storage unit 118 supports encryption and access control measures to protect stored data from unauthorized access or tampering. The cloud storage unit 118 integrates with distributed storage architectures, enabling scalability and high availability of stored data. The cloud storage unit 118 synchronizes with the server 114 to update stored data periodically or in real-time based on transmission schedules. The cloud storage unit 118 enables data-sharing capabilities, allowing secure access to classified driving patterns by authorized entities such as regulatory agencies, insurance providers, or research institutions. The cloud storage unit 118 stores additional contextual information, including weather conditions, road infrastructure details, and external sensor inputs, enhancing the accuracy of behavioral analysis. The cloud storage unit 118 facilitates long-term data retention policies, complying with data governance and privacy regulations. The cloud storage unit 118 provides data analytics functionalities, enabling trend analysis, anomaly detection, and predictive modeling based on stored classified driving patterns. The cloud storage unit 118 supports integration with third-party applications, allowing interoperability with existing data management frameworks and analytical tools. The cloud storage unit 118 optimizes storage utilization through compression techniques, data deduplication, and intelligent caching mechanisms, reducing storage overhead while maintaining retrieval efficiency.
In an embodiment, the server 114 may extract multiple contextual factors, including a time of day, a weather condition, and a traffic density, using data obtained from external application programming interfaces (APIs) and the plurality of sensors 102. The time of day is derived from system timestamps, allowing differentiation between daytime and nighttime driving behaviors. The weather condition is determined by retrieving meteorological data from online weather services, which comprise parameters such as temperature, humidity, precipitation, and visibility. The traffic density is assessed using road congestion data obtained from navigation services, traffic monitoring systems, and vehicle movement patterns captured by the global positioning system (GPS) unit 108. The server 114 correlates the extracted contextual factors with the classified driving patterns to identify specific behavioral characteristics of a driver. The correlation process involves associating abrupt accelerations, braking events, and lane changes with traffic congestion levels, identifying cautious or aggressive driving tendencies under varying weather conditions, and detecting differences in speed regulation based on time-dependent traffic flow. The server 114 applies statistical models to establish relationships between contextual factors and driving patterns, enabling the generation of insights related to risk assessment and behavioral trends.
In an embodiment, the server 114 may identify a specific stress-inducing condition by analyzing abrupt deviations in the driving patterns correlated with contextual data and provides stress-reducing feedback through the external display device 116. The stress-inducing condition is detected by monitoring irregular acceleration, sudden braking, rapid lane switching, or erratic speed fluctuations captured by the plurality of sensors 102. The correlation with contextual data comprises evaluating road congestion, adverse weather, and proximity to other vehicles using external data sources and onboard sensors. The server 114 applies threshold-based analysis to determine significant deviations from normal driving behavior, distinguishing between voluntary manoeuvres and reactions influenced by external stress factors. The classification of stress-inducing conditions is based on predefined categories such as traffic-induced stress, environmental stress, and fatigue-related stress. The external display device 116 presents real-time notifications or visual indicators to inform the driver of detected stress-inducing conditions. The feedback mechanisms comprise displaying relaxation prompts, suggesting alternative routes with reduced congestion, or activating adaptive vehicle control features such as smooth acceleration guidance. The server 114 adapts stress assessment criteria dynamically based on historical driving data and user-specific behavior trends.
In an embodiment, the server 114 may identify patterns indicating a drowsiness status by analyzing a prolonged steady driving behavior combined with contextual data and alerts the driver through the external display device 116. The detection of drowsiness is based on continuous monitoring of vehicle control inputs, including prolonged periods of constant speed, delayed response in steering adjustments, and lack of acceleration or braking variations recorded by the plurality of sensors 102. The contextual data comprises time of day, recent driving duration, and environmental conditions that contribute to driver fatigue. The server 114 processes data related to head movements, eye closure patterns, and posture changes captured by the camera 110 to supplement behavioral indicators of drowsiness. The external display device 116 generates visual or auditory alerts when drowsiness-related patterns exceed predefined thresholds, prompting the driver to take corrective actions such as taking rest breaks or adjusting vehicle settings for enhanced alertness. The server 114 evaluates past driving records to establish personalized drowsiness detection parameters customized to individual driving habits. The server 114 integrates drowsiness-related insights with real-time navigation data, recommending rest stops or adjusting route plans based on detected fatigue levels.
In an embodiment, the server 114 may compare the classified driving patterns with regional driving norms stored in the cloud storage unit 118 to identify deviations and provide a region-specific driving behavior assessment. The regional driving norms comprise speed limits, typical acceleration and braking profiles, lane discipline regulations, and other standard driving practices applicable to a specific geographical location. The comparison process involves analyzing deviations in speed regulation, lane-changing frequency, and intersection approach behavior relative to locally established norms. The server 114 retrieves regional driving norms from regulatory databases, transportation authority records, and historical driving data aggregated from multiple users. The deviations from regional norms are categorized based on severity levels, ranging from minor variations to high-risk behavioral anomalies. The external display device 116 presents assessments of compliance with regional norms through graphical indicators or numerical scoring metrics. The server 114 dynamically updates regional driving norms based on evolving traffic regulations, infrastructure developments, and localized driving trends. The server 114 stores region-specific assessments in the cloud storage unit 118, enabling long-term tracking of behavioral adaptation to different driving environments. The server 114 facilitates integration with traffic management authorities by providing aggregated compliance reports for policy analysis and road safety improvement initiatives.
In an embodiment, the server 114 may analyze transitions between the classified driving patterns to identify potential triggers for aggressive behavior and suggests mitigation strategies through the external display device 116. The analysis of transitions involves detecting sequential changes in acceleration, braking intensity, lane-switching frequency, and throttle control variations captured by the plurality of sensors 102. The server 114 identifies patterns where smooth driving behavior shifts toward aggressive tendencies, determining the underlying triggers such as sudden congestion, proximity to aggressive drivers, or environmental stressors like poor visibility. The classification of aggressive behavior transitions comprises identifying rapid acceleration bursts, sharp braking sequences, and frequent overtaking manoeuvres. The server 114 assigns contextual labels to transitions, distinguishing between reactive aggression caused by external conditions and habitual aggressive driving tendencies. The external display device 116 provides real-time feedback to the driver through visual or auditory cues, suggesting behavioral adjustments such as maintaining safe following distances, moderating acceleration inputs, and avoiding frequent lane changes. The server 114 personalizes mitigation strategies based on historical driving records, adapting intervention thresholds to individual driver profiles.
In an embodiment, the server 114 may predict collision risks by analyzing the classified driving patterns, the contextual factors, and the proximity data gathered by the plurality of sensors 102. The classified driving patterns comprise acceleration behavior, braking tendencies, lane-changing frequency, and turn execution characteristics, which provide insight into risk scenarios. The contextual factors comprise traffic conditions, road layout, weather variations, and time-dependent congestion levels retrieved from external application programming interfaces and onboard sensor inputs. The proximity data comprises the relative position, distance, and movement patterns of surrounding vehicles, pedestrians, and static obstacles, captured by the global positioning system unit 108, the camera 110, and additional environmental monitoring sensors. The server 114 applies threshold-based assessment techniques to determine whether the detected proximity conditions and driving behaviors indicate an elevated risk of collision. The analysis comprises identifying sudden deceleration events, lane encroachments, unexpected object appearances, and failure to maintain a safe following distance. The external display device 116 provides real-time alerts to inform the driver of impending collision risks and suggests corrective actions such as adjusting speed, modifying lane position, or maintaining a greater buffer distance
In an embodiment, the server 114 may generate anonymized driving pattern reports for urban planning applications and traffic management applications, based on aggregated data stored in the cloud storage unit 118. The anonymized driving pattern reports comprise statistical analyses of vehicle movement trends, congestion hotspots, acceleration and deceleration behaviors, and common route preferences without associating data with individual users. The aggregation process involves collecting classified driving behaviors from multiple vehicles, filtering identifiable information, and generating generalized insights that reflect traffic dynamics at a regional or city-wide scale. The urban planning applications comprise optimizing road network designs, improving traffic signal placement, and assessing the need for infrastructure modifications such as additional lanes, speed regulation measures, or pedestrian safety enhancements. The traffic management applications comprise evaluating peak-hour congestion patterns, identifying accident-prone intersections, and optimizing real-time traffic control strategies. The anonymized reports facilitate collaboration with municipal authorities, transportation agencies, and research organizations for evidence-based policymaking. The server 114 periodically updates reports using continuously acquired driving pattern data, affirming relevance to evolving traffic conditions. The cloud storage unit 118 retains historical driving pattern records, allowing trend analysis over extended periods. The external display device 116 presents summarized insights for review by urban planners, traffic engineers, and transportation policymakers.
In an embodiment, the server 114 may analyze correlations between aggressive driving patterns and a vehicle fuel consumption, providing eco-driving recommendations through the external display device 116. The aggressive driving patterns comprise frequent rapid acceleration, harsh braking, excessive idling, and abrupt steering adjustments captured by the plurality of sensors 102. The vehicle fuel consumption data is derived from engine performance parameters, fuel injection rates, and energy expenditure metrics, retrieved from onboard diagnostics and external data sources. The server 114 establishes statistical relationships between aggressive driving tendencies and increased fuel usage, identifying specific behaviors that contribute to inefficient fuel consumption. The eco-driving recommendations comprise suggestions for smoother acceleration control, gradual braking application, optimized cruising speeds, and route selection strategies to minimize fuel wastage. The external display device 116 provides real-time feedback, alerting the driver to fuel-inefficient manoeuvres and displaying corrective measures. The recommendations are customized based on historical driving behavior, real-time traffic conditions, and external environmental factors. The server 114 adjusts eco-driving guidance dynamically, considering variations in terrain, vehicle load, and prevailing weather conditions.
FIG. 2 illustrates a method 200 for analyzing and classifying driving patterns of a two-wheeler, in accordance with the embodiments of the present disclosure. At step 202, data related to speed, acceleration, orientation, location, and visual driving conditions of a two-wheeler is gathered using a plurality of sensors 102. The plurality of sensors 102 comprises an accelerometer 104 for measuring linear acceleration, a gyroscope 106 for detecting angular velocity, a global positioning system unit 108 for acquiring location coordinates, and a camera 110 for capturing visual data of the surrounding environment. The accelerometer 104 detects changes in speed and movement patterns, while the gyroscope 106 determines the tilting and rotation of the two-wheeler. The global positioning system unit 108 provides continuous tracking of position, speed, and direction of travel of the vehicle. The camera 110 records images or videos of road conditions, traffic signals, lane markings, and obstacles. The plurality of sensors 102 continuously monitors driving dynamics and transmits raw sensor readings for further processing. The gathered data comprises real-time measurements and historical driving records, allowing identification of behavioral trends and movement characteristics. The data is temporarily stored in a local memory before being transmitted for classification and analysis.
At step 204, the gathered data from the plurality of sensors 102 is transmitted to a server 114 through a communication interface 112. The communication interface 112 facilitates data exchange between the plurality of sensors 102 and the server 114 using wired or wireless communication methods. The communication interface 112 transmits sensor data packets at predefined intervals or in real time based on system requirements. The transmission methods comprise Bluetooth, Wi-Fi, cellular networks, or direct wired connections, depending on the infrastructure and network availability. The communication interface 112 encodes and encrypts data to maintain transmission security and prevent unauthorized access. The data packets comprise timestamped sensor readings, allowing synchronization of measurements from different sensors. The transmission parameters are dynamically adjusted based on network conditions, affirming stable data transfer without packet loss. The server 114 receives the transmitted data and prepares the received data for classification and pattern analysis. The communication interface 112 supports bidirectional communication, allowing the server 114 to send acknowledgment signals or request additional data from the plurality of sensors 102 as needed. The transmitted data is temporarily stored in a buffer memory at the server 114 before undergoing further processing.
At step 206, the transmitted data is received at the server 114 for further processing and classification. The server 114 collects raw sensor data from multiple sources and organizes the received data into structured formats. The server 114 applies preliminary filtering to remove duplicate entries, incomplete records, or sensor anomalies that may affect the classification process. The received data is time-synchronized to enable proper correlation between acceleration, orientation, speed, and visual driving conditions. The server 114 verifies the integrity of the transmitted data by checking for inconsistencies or missing values. The received data is stored in a database or temporary storage for classification and analysis. The server 114 manages data processing workloads, distributing tasks based on system resources and computational capacity. The server 114 continuously monitors incoming data streams and dynamically adjusts processing priorities based on transmission frequency and real-time analysis requirements. The received data is formatted into a structured dataset, allowing efficient clustering and behavioral classification in subsequent steps. The server 114 performs data aggregation when multiple data streams are received from different sources, enabling analysis of driving patterns.
At step 208, a clustering technique is applied to categorize the received data into multiple clusters. The clustering technique groups data points based on similarity measures such as acceleration patterns, speed variations, lane change frequencies, and braking intensity. The server 114 analyzes sensor data to identify behavioral patterns associated with different driving conditions. The clustering technique assigns sensor readings to distinct groups, facilitating the classification of driving behavior. The clustering technique utilizes distance-based, density-based, or hierarchical methods to form meaningful clusters. The clusters represent different driving styles, including smooth driving, aggressive acceleration, abrupt braking, or erratic lane-switching behavior. The server 114 refines cluster boundaries dynamically as additional data is processed, improving classification accuracy over time. The clustering technique allows identification of common trends in driving behavior based on recurring movement characteristics. The generated clusters serve as input for classification models that further analyze and categorize driving patterns. The clustering process is periodically updated with newly received data to reflect real-time driving conditions. The server 114 optimizes cluster formation by eliminating redundant groupings and assuring that each cluster accurately represents distinct driving behaviors. The clustering results are stored for classification and pattern recognition in subsequent steps.
At step 210, at least one supervised machine learning model is executed to analyze each cluster and classify the driving patterns into categories comprising aggressive behaviors and non-aggressive behaviors. The supervised machine learning model is selected from a group consisting of Random Forest classifier, XGBoost, and K-nearest neighbor. The supervised machine learning model utilizes labeled training datasets containing pre-classified driving behaviors to improve classification accuracy. The server 114 processes the clustered sensor data and extracts relevant features such as acceleration magnitude, braking force, lane position stability, and throttle input variations. The supervised machine learning model assigns each driving pattern to predefined behavioral categories, distinguishing between safe and risky driving tendencies. The classification process considers environmental factors, road conditions, and time-dependent variations to improve accuracy. The classified results are evaluated against historical driving data to detect anomalies or behavioral deviations. The server 114 continuously updates classification parameters based on newly gathered data, refining behavioral categorization for improved decision-making. The classification output is stored in a structured format for visualization and further analysis. The classified driving patterns provide insights into driver habits, risks, and overall road safety conditions.
At step 212, the classified driving patterns, a traffic congestion, and the road conditions are stored in the cloud storage unit 118 operatively connected to the server 114. The classified driving patterns comprise categorized records of acceleration tendencies, braking behaviors, lane discipline, and speed consistency. The traffic congestion data comprises real-time and historical congestion trends derived from vehicle density analysis, road occupancy measurements, and traffic flow patterns. The road conditions comprise surface quality indicators, obstacle detection records, and environmental parameters affecting driving stability. The cloud storage unit 118 organizes stored data using indexed structures, allowing efficient retrieval and further processing. The storage format comprises structured databases, time-series logs, and analytical datasets to facilitate long-term monitoring. The cloud storage unit 118 applies access control policies to assure data security and privacy compliance. The classified data is periodically updated based on new observations from the plurality of sensors 102. The stored records are accessible for real-time analytics, regulatory assessments, and driving behavior improvements. The cloud storage unit 118 maintains a backup of stored data, assuring availability for future reference. The stored driving patterns contribute to predictive modeling and decision support for road safety applications.
At step 214, the classified driving pattern is provided to at least one external display device 116 for visualization and user interaction. The external display device 116 presents classified driving behaviors through graphical dashboards, numerical indicators, or real-time alerts. The external display device 116 is implemented as a digital dashboard, mobile application, infotainment system, or heads-up display. The external display device 116 provides feedback on acceleration smoothness, braking consistency, and lane discipline, allowing users to assess driving habits. The external display device 116 generates alerts for aggressive driving behaviors, prompting corrective actions. The external display device 116 allows access to stored historical records, enabling trend analysis and behavioral tracking. The external display device 116 supports customization options for displaying selected driving metrics, notification preferences, and visual formats. The external display device 116 synchronizes with the cloud storage unit 118 to access updated driving behavior insights. The external display device 116 supports real-time updates, making sure continuous monitoring of driving classification outcomes. The external display device 116 facilitates integration with third-party safety monitoring applications, insurance platforms, and regulatory frameworks. The external display device 116 provides structured driving reports summarizing behavioral classifications, enhancing decision-making for drivers and traffic management authorities.
FIG. 3 illustrates a sequence diagram of a system 100 for analyzing and classifying the driving patterns of a two-wheeler, in accordance with the embodiments of the present disclosure. The system 100 comprises a plurality of sensors 102, a server 114, a cloud storage unit 118, and external display devices 116. The plurality of sensors 102, including an accelerometer 104, a gyroscope 106, a global positioning system unit 108, and a camera 110, gathers and transmits driving data to the server 114. The server 114 receives the transmitted data and categorizes the data into clusters based on driving behaviors, environmental conditions, and road characteristics. The server 114 further processes the clustered data by executing at least one machine learning model to classify driving patterns into predefined categories, including aggressive behaviors and non-aggressive behaviors. The classified driving patterns, along with associated traffic congestion levels and road conditions, are stored in the cloud storage unit 118 for historical analysis and future reference. The server 114 provides the classified driving patterns to external display devices 116 for visualization and driver feedback. The external display devices 116 present categorized driving behaviors, alerts, and recommendations, enabling users to assess and modify driving habits. The cloud storage unit 118 allows retrieval of stored data for long-term monitoring and regulatory analysis.
In an embodiment, a plurality of sensors 102 gathers data related to speed, acceleration, orientation, location, and visual driving conditions of a two-wheeler, allowing continuous monitoring of movement characteristics. An accelerometer 104 detects linear acceleration, enabling identification of sudden speed changes. A gyroscope 106 measures angular velocity, allowing determination of tilt and rotational movement. A global positioning system unit 108 provides real-time positioning, affirming accurate tracking of routes, speed, and travel patterns. A camera 110 captures visual data related to road conditions, traffic signals, and obstacles, contributing to enhanced situational awareness. The plurality of sensors 102 operates in combination to create a dataset for driving pattern classification. The continuous data acquisition from multiple sensor sources allows detailed behavioral analysis, identifying variations in acceleration control, braking response, and lane discipline. The gathered data is processed to detect deviations from normal driving behavior, providing insights into movement stability and external environmental influences.
In an embodiment, a communication interface 112 transmits gathered data from the plurality of sensors 102 to a server 114 for processing and classification. The communication interface 112 supports wired or wireless transmission, enabling reliable data exchange between the sensing unit and processing components. The transmission process comprises time-stamped sensor readings, maintaining synchronization between multiple data sources. The communication interface 112 dynamically adjusts data transfer rates based on network conditions, reducing latency and minimizing packet loss. The communication interface 112 facilitates continuous or batch transmission depending on real-time processing requirements. The implementation of secure transmission mechanisms, including encryption and authentication techniques, prevents unauthorized access and data manipulation. The communication interface 112 enables bidirectional communication, allowing the server 114 to request additional data when required, affirming completeness in transmitted datasets. The transmission of structured sensor data allows efficient categorization and behavioral assessment based on received inputs.
In an embodiment, the server 114 receives transmitted data from the plurality of sensors 102 and applies a clustering technique to categorize received data into multiple clusters based on similarities in driving behavior. The clustering technique groups acceleration trends, braking sequences, lane-change patterns, and speed variations to create structured representations of driving habits. The server 114 analyzes statistical relationships within the collected dataset to identify recurring movement trends. The clustering technique provides segmented datasets, assuring classification accuracy when determining driving behavior categories. The clusters represent distinct driving styles, including controlled movements, abrupt accelerations, and erratic braking sequences. The categorized data is used as an input for classification models that process behavioral insights for further analysis. The clustering technique dynamically adapts to newly acquired data, refining group definitions based on evolving driving trends.
In an embodiment, the server 114 executes at least one supervised machine learning model selected from a group consisting of Random Forest classifier, XGBoost, and K-nearest neighbor to analyze each cluster and classify driving patterns into categories comprising aggressive behaviors and non-aggressive behaviors. The server 114 extracts key features from categorized datasets, identifying acceleration intensity, braking consistency, throttle response, and lane discipline to determine driving tendencies. The classification process applies predefined behavioral thresholds, distinguishing between normal and high-risk driving behaviors. The supervised machine learning model refines classification accuracy through iterative processing and historical data comparison. The classification results assist in detecting patterns associated with erratic movements, abrupt stops, or unsafe lane transitions. The behavioral classification contributes to personalized driving assessments, enabling identification of long-term driving habits. The classification outcomes undergo validation using reference datasets, enhancing reliability in behavior differentiation. The classification framework updates dynamically, incorporating new sensor observations and contextual variables to enhance detection precision.
In an embodiment, the server 114 provides the classified driving patterns to one or more external display devices 116 for real-time visualization and driver feedback. The external display devices 116 present categorized driving behaviors through graphical dashboards, text-based alerts, and trend analysis charts. The external display devices 116 indicate deviations from optimal driving behaviors, allowing users to assess and adjust movement characteristics accordingly. The external display devices 116 generate alerts for high-risk driving tendencies, notifying the driver about safety concerns. The external display devices 116 facilitate review of historical driving records, enabling retrospective analysis of behavioral trends. The external display devices 116 integrate with cloud-based storage to retrieve stored classification results and past driving assessments. The external display devices 116 support customization options, allowing adjustment of displayed metrics, alert sensitivities, and reporting formats.
In an embodiment, a cloud storage unit 118 stores classified driving patterns, traffic congestion data, and road conditions for long-term reference and predictive analysis. The cloud storage unit 118 maintains structured datasets, allowing retrieval and examination of driving behavior trends. The cloud storage unit 118 categorizes stored records based on time, location, and behavioral classifications, facilitating efficient search and analysis. The cloud storage unit 118 incorporates security mechanisms, including encrypted storage and access control policies. The cloud storage unit 118 dynamically updates stored data, incorporating new classification results and contextual observations. The cloud storage unit 118 enables integration with external regulatory systems, allowing access to historical behavior assessments for compliance verification and safety evaluations. The cloud storage unit 118 retains backup copies of stored data, preventing information loss due to unexpected failures or connectivity interruptions. The stored data contributes to long-term behavioral analytics, facilitating predictive assessments and adaptive classification improvements.
In the description of the present invention, it is also to be noted that, unless otherwise explicitly specified or limited, the terms “disposed,” “mounted,” and “connected” are to be construed broadly, and may for example be fixedly connected, detachably connected, or integrally connected, either mechanically or electrically. They may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Modifications to embodiments and combination of different embodiments of the present disclosure described in the foregoing are possible without departing from the scope of the present disclosure as defined by the accompanying claims. Expressions such as “comprising”, “comprising”, “incorporating”, “have”, “is” used to describe and claim the present disclosure are intended to be construed in a non- exclusive manner, namely allowing for items, components or elements not explicitly described also to be present. Reference to the singular is also to be construed to relate to the plural where appropriate.
Although embodiments have been described with reference to a number of illustrative embodiments thereof, it should be understood that numerous other modifications and embodiments can be devised by those skilled in the art that will fall within the spirit and scope of the principles of this disclosure. More particularly, various variations and modifications are possible in the component parts and/or arrangements of the subject combination arrangement within the scope of the present disclosure, the drawings and the appended claims. In addition to variations and modifications in the component parts and/or arrangements, alternative uses will also be apparent to those skilled in the art.
,CLAIMS:WE CLAIM:
1. A system 100 for analyzing and classifying the driving patterns of a two-wheeler, the system 100 comprising:
a plurality of sensors 102, wherein the sensors 102 include an accelerometer 104, a gyroscope 106, a global positioning system (GPS) unit 108, and a camera 110, operable to gather data related to speed, acceleration, orientation, location, and the visual driving conditions of the two-wheeler;
a communication interface 112 configured to transmit data, gathered by the plurality of sensors 102, to a server 114;
the server 114 is configured to:
o receive transmitted data;
o apply a clustering technique to categorize received data into multiple clusters;
o execute at least one supervised machine learning model selected from a group consisting of Random Forest classifier, XGBoost, and K-nearest neighbor for analysing each cluster to classify the driving patterns into the categories, wherein the categories comprise the aggressive behaviors and the non-aggressive behaviors; and
o provide the classified driving patterns to one or more external display devices 116;
a cloud storage unit 118 operatively connected to the server 114, wherein the cloud storage unit 118 stores the classified driving patterns, a traffic congestion and the road conditions.
2. The system 100 of claim 1, wherein the server 114 is further operable to extract the contextual factors, including a time of day, a weather condition, and a traffic density, using data obtained from the external application programming interfaces (APIs) and the plurality of sensors 102, and to correlate the extracted contextual factors with the classified driving patterns to identify the specific behavioral characteristics of a driver.

3. The system 100 of claim 1, wherein the server 114 is further operable to identify a specific stress-inducing condition by analyzing the abrupt deviations in the driving patterns correlated with contextual data, and to provide a stress-reducing feedback through the external display device 116.

4. The system 100 of claim 1, wherein the server 114 identifies the patterns indicating a drowsiness status by analyzing a prolonged steady driving behavior combined with contextual data and the alerts for the driver through the external display device 116.

5. The system 100 of claim 1, wherein the server 114 is further operable to compare the classified driving patterns with the regional driving norms stored in the cloud storage unit 118 to identify the deviations and provide a region-specific driving behavior assessment.

6. The system 100 of claim 1, wherein the server 114 analyzes the transitions between the classified driving patterns to identify the potential triggers for aggressive behavior and suggests the mitigation strategies through the external display device 116.

7. The system 100 of claim 1, wherein the server 114 predicts the potential collision risks by analyzing the classified driving patterns, the contextual factors, and the proximity data gathered by the plurality of sensors 102.

8. The system 100 of claim 1, wherein the server 114 is further operable to generate the anonymized driving pattern reports for the urban planning applications and the traffic management applications, based on aggregated data stored in the cloud storage unit 118.
9. The system 100 of claim 1, wherein the server 114 analyzes the correlations between the aggressive driving patterns and a vehicle fuel consumption, providing the eco-driving recommendations through the external display device 116.

10. A method 200 for analyzing and classifying driving patterns of a two-wheeler, the method 200 comprising:
gathering, data related to speed, acceleration, orientation, location, and the visual driving conditions of the two-wheeler using a plurality of sensors 102, wherein the sensors 102 include an accelerometer 104, a gyroscope 106, a global positioning system (GPS) unit 108, and a camera 110;
transmitting, the gathered data from the plurality of sensors 102 to a server 114 through a communication interface 112;
receiving, the transmitted data at the server 114;
applying, a clustering technique on the received data to categorize into multiple clusters;
executing, at least one supervised machine learning model, selected from a group consisting of Random Forest classifier, XGBoost, and K-nearest neighbor, to analyze each cluster and classify the driving patterns into the categories comprising the aggressive behaviors and the non-aggressive behaviors;
storing, the classified driving patterns, a traffic congestion, and the road conditions, in the cloud storage unit 118 operatively connected to the server 114; and
providing, the classified driving pattern to at least one external display device 116.