Description:FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
The Patent Rules, 2003
COMPLETE SPECIFICATION
(See sections 10 & rule 13)
1. TITLE OF THE INVENTION
A SYSTEM AND METHOD FOR REAL-TIME COLLISION PREDICTION AND AVOIDANCE IN VEHICULAR AD HOC NETWORKS (VANETs)
2. APPLICANT (S)
S. No. NAME NATIONALITY ADDRESS
1 NMICPS Technology Innovation Hub On Autonomous Navigation Foundation IN C/o Indian Institute of Technology Hyderabad, Kandi, Sangareddy, Telangana– 502284, India.
2 Indian Institute Of Technology Hyderabad IN Kandi, Sangareddy, Telangana– 502284, India.
3. PREAMBLE TO THE DESCRIPTION
COMPLETE SPECIFICATION

The following specification particularly describes the invention and the manner in which it is to be performed.

FIELD OF INVENTION:
[001] The present invention relates to the field of artificial intelligence system. The present invention in particular relates to a system and method for real-time collision prediction and avoidance in vehicular ad hoc networks (VANETs).
DESCRIPTION OF THE RELATED ART:
[002] Vehicles may be equipped with collision avoidance systems configured to detect and avoid objects in an operating environment. The objects may include mobile objects, such as other vehicles, cyclists, pedestrians, etc., Traditional collision avoidance systems may avoid collisions by simply identifying the presence of surfaces in an environment and adjusting a velocity of a vehicle to avoid collision with a surface. However, these traditional systems may cause the vehicle to yield in situations in which it is unnecessary and unnatural, thereby potentially causing traffic delays.
[003] Reference may be made to the following:
[004] IN Publication No. 202321053515 relates to an innovative system and method for enhancing the security and reliability of vehicular Ad-Hoc networks (VANETs) through the utilization of artificial intelligence (AI)-based attack prediction. VANETs, being a cornerstone of modern intelligent transportation systems, facilitate seamless communication and data exchange among vehicles and infrastructure, promising enhanced road safety, traffic efficiency, and passenger experience. However, the openness and dynamism of VANETs expose them to an array of security threats, ranging from data manipulation to denial-of-service attacks, jeopardizing their integrity and effectiveness. The proposed system employs advanced AI techniques and machine learning algorithms to predict, identify, and mitigate potential security attacks within the VANET environment. The system comprises a multi-faceted architecture, encompassing data collection, feature extraction, machine learning, attack prediction, and alert/mitigation modules.
[005] IN Publication No. 202327048256 relates to systems and methods performed by vehicle-to-everything (V2X) system participant to determine whether a misbehavior condition may have occurred based on the generation and/or receipt of a V2X message. The detection of a misbehavior condition may occur if the V2X message is generated and/or received too frequently or not frequently enough. In addition, a misbehavior condition may be detected if the generated and/or received V2X message does include the appropriate security credential.
[006] IN Publication No. 201741023965 relates to intelligent systems are in used with every aspect of systems, vehicles are the critical systems which are real time and lives are involved. Impact detection and notification is also one of the life saving and critical information provider system. The objective of this project is to efficiently avoid the accident of vehicles and to provide a greatest security to the users in adverse. Traffic accidents keep with a yearly increasing of a high rate. This system are advanced technological innovations are available today for vehicle safety. LDR Sensor & Speed sensor detection are the vital andof great importance from the perspective of passenger safety and traffic safety. The system takes over the throttle of the car to maintain a steady speed as set by the driver. In our proposed method the relative speed and distance of all the vehicles around a particular vehicle is estimated using speed sensors based on those results the speed of that particular vehicle is controlled to avoid early accidents. In road unit consist of LDR sensor, microcontroller, wireless transceiver.
[007] IN Publication No. 1486/CHE/2004 relates to wireless communication and specifically to ad hoc wireless networks. More particularly, this invention relates to methods for admission control and scheduling in an ad hoc wireless network for different classes of flows Cl, C2 and C3 where said admission control mechanism provide efficient CIoS by evaluating the probability of QoS violation and admit those calls whose probability of iaoS violation is below a specified threshold\ Ivtierein Admission Control for Class Ci evaluates the probability as where destination node is a one-hop neighbour of the source node; Admission Control for Class Cl evaluates the probability as where source and destination nodes are not one hop neighbours; Admission Control for Class C2 and class C3 flows evaluates the average or mean delay requirement and a requirement on the rate; Scheduling class Cl flows when admitted flows violate the QoS requirements; and Scheduling class C2 and class C3 flows.
[008] Publication No. US2024149868 relates to a vehicle computing system may implement techniques to control a vehicle to avoid collisions between the vehicle and agents (e.g., dynamic objects) in an environment. The techniques may include generating a representation of a path of the vehicle through an environment as a polygon. The vehicle computing system may compare the two-dimensional path with a trajectory of an agent determined using sensor data to determine a collision zone between the vehicle and the agent. The vehicle computing system may determine a risk of collision based on predicted velocities and probable accelerations of the vehicle and the agent approaching and traveling through the collision zone. Based at least in part on the risk of collision, the vehicle computing system may cause the vehicle to perform an action.
[009] Patent No. US9340154 relates to a method and nodes for collision avoidance in a vehicular network comprising a plurality of vehicles and lower-speed users. Data from a plurality of position notification messages sent from the lower-speed-node is received (directly from the lower-speed-node or through a road-side-node) and processed, in a vehicle-on-board-node of a moving vehicle, with positioning data of the moving vehicle into collision avoidance instructions. The collision avoidance instructions may be provided inside the vehicle by displaying the instructions on a screen and/or broadcasting the instructions through a speaker in the vehicle. The collision avoidance instructions may also be automatically applied by modifying dynamic parameters of the vehicle (e.g., affecting the vehicle speed and/or changing direction of the vehicle). The position notification message may comprise a sequential number value; a temporary ID value; a geographic latitude value; a geographic longitude value; and a class-speed value.
[010] Patent No. US9361802 relates to mesh node modules are associated with vehicles and companion nodes can dynamically form a mesh network which uploads location information of the nodes and in some cases additional information+, e.g., road condition or proximity to objects.
[011] Publication No. CN112109704 relates to a vehicle collision avoidance dynamic safety path planning method based on accurate trajectory prediction. The method comprises the steps of obtaining vehicle state information and parameter information of a vehicle and an obstacle vehicle; constructing a vehicle trajectory prediction model based on a three-degree-of-freedom vehicle dynamics model and a long-short-term memory recurrent neural network LSTM; using the vehicle trajectory prediction model to predict the driving trajectories of the vehicle and the obstacle vehicle, and obtaining the driving trajectory prediction data of the vehicle and the obstacle vehicle; constructing a vehicle dynamic safety path planning model by fusing mesh generation, an artificial potential field method and a high-order polynomial curve fitting method; based on the driving trajectory prediction data of the vehicle and the obstacle vehicle obtained in the step 3, using the vehicle dynamic safety path planning model to obtain an optimal collision avoidance path of the vehicle; and carrying out speed and acceleration real-time matching on the optimal collision avoidance path obtained in the step 5. Compared with the prior art, the method has the advantages of high precision, safety, feasibility, comfort and the like.
[012] Publication No. US2018032891 relates to real-time collision avoidance in moving vehicles includes initializing a prior collision distribution from a manufacturer's vehicle calibration, receiving driver data acquired from a driver when a vehicle is driven and vehicular data acquired from the vehicle being driven by the driver, determining a conditional collision probability using features derived from the driver data and the vehicular data and a model for the driver, calculating posterior probability collision distribution from the conditional collision probability and the prior collision distribution, and determining a probability of a collision occurring from the posterior probability collision.
[013] Publication No. CN111556464 relates to a distributed Internet-of-vehicles MAC layer merging collision prediction and avoidance method based on a TDMA technology. A vehicle uses a TDMA communication technology to realize Internet-of-vehicles communication, and the vehicle collects and updates driving states and time slot use information of the vehicle and a one-hop neighbor in real time, embeds the driving states and the time slot use information into a beacon frame, and periodically sends the beacon frame to the one-hop neighbor vehicle to complete information exchange; and the vehicle signal transceiving equipment executes potential collision detection according to the received information, judges whether the vehicle signal transceiving equipment is in a dangerous distance, executes potential collision prediction and judges whether collision conditions are met or not if the vehicle signal transceiving equipment is in the dangerous distance, and executes potential collision elimination if yes.
[014] Publication No. CN117333843 relates to collision prediction method and device, the storage medium and the computer equipment provided by the invention, for the current simulation moment, the obstacle trajectory and the multiple target vehicle prediction trajectories are determined, so that the vehicle prediction position on each target vehicle prediction trajectory and the current position of the obstacle on the obstacle trajectory are determined; therefore, the collision probability of the vehicle and the obstacle on each target vehicle prediction track can be calculated according to the plurality of vehicle prediction positions and the current position of the obstacle, and based on the collision probability of the vehicle on each target vehicle prediction track and the pre-acquired occurrence probability of each target vehicle prediction track, the collision probability of the vehicle on each target vehicle prediction track can be calculated.
[015] Publication No. CN113094808 relates to an automobile collision damage grade real-time prediction method based on simulation data and artificial intelligence, which comprises the following steps: collecting a vehicle collision video, recording data, and establishing a car model and a big truck model; acquiring a simulation data set of a vehicle collision condition required by the deep learning neural network; establishing a deep convolutional neural network, and initializing the established deep convolutional neural network; training a deep convolutional neural network based on the obtained simulation data set; and based on the trained deep convolutional neural network, predicting the collision damage degree of the vehicle and the member in a short time in the future in real time.
[016] Publication No. US2005060069 relates to apparatus for sensing a potential rollover situation involving a vehicle including an inertial reference unit including three accelerometers and three gyroscopes which provide data on vehicle motion, vehicle control devices arranged to affect control of the vehicle and a processor coupled to the inertial reference unit and the vehicle control devices. The processor includes an algorithm arranged to receive data from the inertial reference unit and control the vehicle control devices to apply the throttle, brakes and steering to prevent the rollover, optionally in consideration of the position of the vehicle as provided by a map database or location determining system.
[017] Publication No. CN117389296 relates to a Markov decision and probability model-based large unmanned aerial vehicle formation collision avoidance method, which comprises the following steps of: constructing a Markov state probability model, designing a two-dimensional collision avoidance logic and a three-dimensional collision avoidance logic, and taking a confidence state output by the probability model as input of the collision avoidance logic, and according to a cost value function designed by a performance index, an optimal strategy is solved by using dynamic programming, and the optimal strategy is used as collision avoidance logic for selecting warning suggestions.
[018] Publication No. CN116654017 relates to an intelligent vehicle collision avoidance decision-making and path planning method and system based on surrounding vehicle trajectory prediction under an emergency working condition, and the method comprises the steps: constructing a physical guidance graph convolutional neural network trajectory prediction model, obtaining the longitudinal and transverse coordinates of a surrounding vehicle prediction trajectory, and a probability density function of longitudinal and transverse coordinate prediction errors; then, carrying out driving safety level prediction evaluation, determining whether the current self-vehicle needs to avoid collision or not, preferentially selecting braking collision avoidance when the current self-vehicle needs to avoid collision, carrying out self-vehicle braking collision avoidance and steering lane-changing collision avoidance risk prediction when the braking collision avoidance condition is not met, and selecting a collision avoidance form with relatively low prediction risk; when steering and lane changing collision avoidance are selected, a collision avoidance path of the vehicle in a future time period is planned through an improved artificial potential field and is optimized.
[019] Publication No. US2023182727 relates to a vehicular collision avoidance system includes a sensor disposed at a vehicle for sensing exterior and forwardly of the vehicle. A processor processes sensor data captured by the sensor to determine the presence of a pedestrian ahead of the vehicle and outside a path of travel of the vehicle. The processor determines a projected path of travel of the pedestrian based on movement of the pedestrian.
[020] Publication No. US2023373477 relates to a collision avoidance support device comprises target detection unit, target type determination unit, relative position determination unit, target track prediction unit, and vehicle track prediction unit, obstacle determination unit. The vehicle track prediction unit is configured to enlarge said width of a vehicle predicted track compared with a case where an enlargement condition is not satisfied when the enlargement condition is satisfied. The enlargement condition is satisfied when the relative position determination unit detects that a target determined to be a pedestrian by the target type determination unit is positioned on a travel lane at least once.
[021] Publication No. CN116901835 relates to automobile collision avoidance, in particular to a vehicle collision avoidance prompting method which comprises the following steps: determining a possible collision area of a vehicle; according to the distance between other vehicles in the area where the collision possibly occurs and the vehicle, the collision is predicted through an anti-collision prediction model; the prediction result is fed back to a driver, and an anti-collision prompt is sent out.
[022] Publication No. CN114265425 relates to a multi-rotor unmanned aerial vehicle formation anti-collision control method, and belongs to the technical field of unmanned aerial vehicles. The method comprises the steps of multi-sensor airspace situation awareness, unmanned aerial vehicle kinematics modeling, nonlinear model control prediction, rolling optimization, collision detection, collision avoidance and the like. The invention provides a novel nonlinear model prediction control method, which can realize flight path planning and collision avoidance of the unmanned aerial vehicle in a dynamic environment. And the real-time performance of algorithm execution is ensured based on the proposed rolling optimization.
[023] Publication No. CN113985875 relates to an artificial potential field unmanned vehicle dynamic path planning method based on a collision prediction model. In a dynamic environment, when an artificial potential field method is applied to unmanned vehicle path planning, response is not timely. Based on the situation in the dynamic environment, a radar sensor collects position and motion state information of an unmanned vehicle and a dynamic obstacle in real time, a collision prediction model is established to judge whether the environment is safe or not, then whether a speed potential field is added or not is determined, so that the unmanned vehicle obtains information needing steering in advance; and the problem that a dynamic obstacle induces the movement direction of the unmanned vehicle is solved by establishing a virtual potential field.
[024] Publication No. CN113759983 relates to an anti-disturbance unmanned aerial vehicle collision avoidance method based on differential flatness. According to the method, the disturbance suffered by the unmanned aerial vehicle is accurately and rapidly learned and estimated by using an incremental Gaussian process, and corresponding compensation is performed according to a prediction result of the disturbance in a trajectory planning and unmanned aerial vehicle control process, so that the unmanned aerial vehicle based on differential flatness control can still realize real-time collision avoidance in the disturbance.
[025] Publication No. CN116301011 relates to a multi-agent efficient formation obstacle avoidance method, and belongs to the field of multi-agent formation control. According to the invention, the risk prediction of individual collision is realized through the intelligent body kinematic model; selecting whether to switch the formation by comprehensively comparing the obstacle avoidance cost of the individual and the overall formation change cost of the formation; by constructing a cost matrix corresponding to the intelligent agent and the target point, real-time optimal distribution of the intelligent agent position is realized by means of a distribution algorithm, the transformation speed is accelerated, and the navigation cost is reduced; through a consistency algorithm and an improved dynamic window method, target guidance is introduced, damage of an obstacle avoidance process to a formation is reduced, local optimum of an individual in the obstacle avoidance process can be effectively prevented, and efficient obstacle avoidance of the individual is realized.
[026] Publication No. CN116118724 relates to road vehicle driving control considering driver characteristics, and particularly relates to a vehicle non-emergency collision avoidance method and system based on long-time trajectory prediction. The information acquisition module is used for acquiring real-time response time of a driver, brake performance parameters of a driving vehicle and road parameters; the trajectory prediction module is used for predicting driving trajectories of the vehicle and surrounding vehicles according to the constructed PINN vehicle long-time trajectory prediction model and the information acquired in the information acquisition module; the collision risk coefficient analysis module is used for obtaining a collision risk coefficient through the predicted driving tracks of the vehicle and the surrounding vehicles and analyzing a threshold risk interval where the collision risk coefficient is located; and calling a corresponding vehicle control scheme according to the threshold risk interval to realize vehicle collision avoidance.
[027] Publication No. CN115923845 relates to an intervention type sharing control method and device for an autonomous vehicle in a forward collision avoidance scene, and relates to the technical field of autonomous vehicle control. Comprising the steps of obtaining vehicle state data and inputting the vehicle state data into a forward collision avoidance control model to obtain an optimal nominal collision avoidance track of an automatic driving vehicle; and steering input data of a driver is obtained, and the sharing control method of the autonomous vehicle in the forward collision avoidance scene is obtained. The vehicle model decoupling method for controlling solution and risk prediction in the high-speed forward collision avoidance scene is provided, the solution efficiency of a linear model and the prediction accuracy of a nonlinear model are both considered, risk pre-estimation in the forward collision avoidance scene is completed, the problem of judgment of critical collision avoidance intervention opportunity is solved, and the risk prediction accuracy is improved.
[028] Publication No. CN115626157 relates to a collision avoidance method and device, a storage medium and computer equipment, and the method comprises the steps: predicting whether an obstacle and a main vehicle will collide in a future simulation frame or not in a current simulation frame according to an obstacle prediction track and a main vehicle prediction track corresponding to the current simulation frame; if the obstacle collides with the main vehicle in the future simulation frame, acquiring a predicted position of a target obstacle when the collision occurs; according to the predicted position of the target obstacle, the real-time speed of the obstacle corresponding to the current simulation frame and the predicted trajectory of the main vehicle, calculating avoidance deceleration corresponding to the current simulation frame; and according to the avoidance deceleration corresponding to the current simulation frame and the real-time obstacle speed corresponding to the current simulation frame, determining the real-time obstacle speed corresponding to the next simulation frame, and entering the next simulation frame. By adopting the scheme of the invention, the efficiency of the simulation result can be improved.
[029] The article entitled “A real-time collision prediction mechanism with deep learning for intelligent transportation system” by Xin Wang; Jing Liu; Tie Qiu; Chaoxu Mu; Chen Chen; Pan Zhou; IEEE Transactions on Vehicular Technology Volume: 69, Issue: 9; September 2020 talks about a novel Rear-end Collision Prediction Mechanism with deep learning method (RCPM), in which a convolutional neural network model is established. In RCPM, the dataset is smoothed and expanded based on genetic theory to alleviate the class imbalance problem. The preprocessed dataset is divided into training and testing sets as the input to train our convolutional neural network model. The experimental results show that compared with the Honda, Berkeley and multi-layer perception neural-network-based algorithms, RCPM effectively improves performance to predict rear-end collisions.
[030] The article entitled “A rear-end collision prediction scheme based on deep learning in the internet of vehicles” by Chen Chen; Hongyu Xiang; Tie Qiu, Cong Wang; Yang Zhou; Victor Chang; Journal of Parallel and Distributed Computing Volume 117; July 2018 talks about a probabilistic model named as CPGN (Collision Prediction model based on GA-optimized Neural Network) for decision-making in the rear-end collision avoidance system is proposed, targeting modeling the impact of important influential factors of collisions on the occurring probability of possible accidents in the Internet of Vehicles (IoV). The decision on how to serve the chauffeur is determined by a typical deep learning model, i.e., the BP neural network through evaluating the possible collision risk with V2I (Vehicle-to-Infrastructure) communication, V2V (Vehicle-to-Vehicle) communication and GPS infrastructure supporting. The proper structure of our BP neural network model is deeply learned with training data generated from VISSIM with multiple influential factors considered. In addition, since the selection of the connection coefficient array and thresholds of the neural network has great randomness, a local optimization issue is readily occurring during the modeling procedure. To overcome this problem and consider the ability to efficiently find out a global optimization, this paper chooses the genetic algorithm to optimize the coefficient array and thresholds of proposed neural network. For the purpose of enhancing the convergence speed of the proposed model, we further adjust the studying rate according to the relationship between the actual and predicated values of two adjacent iterations. Simulation results demonstrate that the proposed collision risk evaluation framework could offer rationale estimations to the possible collision risk in car-following scenarios for the next discrete monitoring interval.
[031] The article entitled “Collision avoidance method using vector-based mobility model in tdma-based vehicular ad hoc networks” by Jung-Hyun Bang and Jung-Ryun Lee; Applied Sciences 10(12):4181; 18 June 2020 talks about a collision avoidance method based on the vehicle mobility prediction model in TDMA-based VANET. The proposed algorithm allocates time-slots of TDMA to avoid access and merging collisions by predicting the mobility of nearby vehicles using control time-slot occupancy information, vehicle ID, hop information, vehicle movement direction, and longitude and latitude of a vehicle. Simulation results show that the proposed algorithm can reduce access and merging collision rates compared with other legacy algorithms, and the performance gain of the proposed algorithm is enhanced in road environments when traffic density is high and where vehicles have high mobility and change their travel directions frequently.
[032] Prior art in collision avoidance generally focuses on sensor-based or radar/lidar technologies without extensive use of real-time network data and mathematical optimization. Hence there needed a real-time processing and higher precision in collision risk assessment and avoidance.
[033] In order to overcome above listed prior art, the present invention aims to provide a system and method for real-time collision prediction and avoidance in vehicular ad hoc networks (VANETs). This system reliance on VANETs and sophisticated data processing distinguishes it from these existing technologies.
OBJECTS OF THE INVENTION:
[034] The principal object of the present invention is to provide a system and method for real-time collision prediction and avoidance in vehicular ad hoc networks (VANETs).
[035] Another object of the present invention is to provide a system and method for real-time collision prediction and avoidance enhance road safety in VANETs.
[036] Still another object of the present invention is to assign priority values by CPAPAM enabling efficient resource allocation and traffic management.
[037] Yet another object of the present invention is to provide a system that offers faster response times due to real-time processing and has higher precision in collision risk assessment and avoidance.
[038] Still another object of the present invention is to provide a system for real-time collision prediction and avoidance which is Scalability across different vehicular environments, beneficial for both populated urban areas and high-speed rural settings.
SUMMARY OF THE INVENTION:
[039] The present invention relates to a system and method for real-time collision prediction and avoidance in vehicular ad hoc networks (VANETs). The system utilizes two novel methodologies: 1. Collision Prediction and Priority Value Assignment Methodology (CPAPAM) and Collision Avoidance Warning Methodology (CAWAM).
[040] This system differentiates itself from existing technologies by: Employing convex optimization techniques to dynamically calculate priority values for each node, a method that is highly efficient and precise compared to traditional rule-based or static collision avoidance systems.
[041] Generating real-time collision avoidance alerts using the recursive least squares method to estimate vehicle speed and distance, enhancing the response time and accuracy over systems that might use simpler, delay-prone calculations.
BREIF DESCRIPTION OF THE INVENTION
[042] It is to be noted, however, that the appended drawings illustrate only typical embodiments of this invention and are therefore not to be considered for limiting of its scope, for the invention may admit to other equally effective embodiments.
[043] Fig. 1 shows block diagram of real-time collision prediction and avoidance system;
[044] Fig. 2 shows flowchart according to the present invention.
DETAILED DESCRIPTION OF THE INVENTION:
[045] The present invention provides a system and method for real-time collision prediction and avoidance in vehicular ad hoc networks (VANETs). The invention is designed to enhance road safety in vehicular networks through two primary components:
1. Collision prediction and priority value assignment methodology (CPAPAM): Calculates priority values for each network node based on real-time vehicle data, using collision risk metrics formulated through convex optimization.
2. Collision avoidance warning methodology (CAWAM): Issues alerts for potential collisions based on dynamic estimates of vehicle speed and distance. This system is crucial for both conventional and autonomous driving vehicles within Intelligent Transportation Systems, providing a scalable solution to reduce collision risks and manage traffic effectively.
[046] The system includes several key components and flow processes: Input Data: Takes real-time data like vehicle speed, direction, and relative distance. CPAPAM: Processes this input to assign priority values using a convex optimization framework. CAWAM: Utilizes calculated data to generate timely collision alerts. Output: Displays the priority values and alerts for action.
[047] Reference may be made to figure 1 which shoes the block diagram of the system. The system comprises the input data block (1) represents the data inputs required by both CPAPA and CAWA, including relative distance, speed, and direction of vehicles, as well as real-time estimated functions F(x) and G(x).
[048] The collision prediction and priority value assignment methodology unit (2), which calculates priority values for each node based on collision risk metrics using convex optimization techniques.
[049] The collision avoidance warning block (3) represents the collision avoidance warning methodology, which generates collision avoidance alerts based on real-time estimates of vehicle speed and distance, employing the recursive least squares method.
[050] Arrows (4) indicate the flow of data and operations between the input data, CPAPA, and CAWA.
[051] The output block (5) denotes the priority values assigned by CPAPA and the collision avoidance alerts generated by CAWA.
[052] Collision Prediction and Priority value Assignment Methodology (CPAPAM)-
a) CPAPAM calculates priority values for each node in the VANET based on a collision metric function.
b) Inputs to CPAPAM include relative distance, speed, and direction of vehicles within the network.
c) CPAPAM employs convex optimization techniques to minimize collision risks and assign priority values accordingly.
d) The methodology normalizes collision metrics to ensure priority values range between 0 and 1.
e) Priority values are continuously updated based on real-time vehicle data.
[053] The CPAPA Methodology is a real-time V2X communication methodology that calculates priority values for each node based on a collision metric function.
[054] The methodology uses convex optimization to calculate priority values based on node collision risk. The methodology considers nodes’ relative distance, speed, and direction and can normalize collision metrics to assign higher priorities to nodes with higher collision risks. The methodology is implemented using the Convex (CVX) optimization technique, making it efficient for realtime priority value calculation for all nodes. The Hessian matrix of the function

[055] for all values of V elrel, Dist, and ? and so is convex, hence it is used here.
[056] The steps of the methodology are listed here:
[057] 1. Define the total number of nodes (vehicles) in VANET, ?V
= Number of nodes
[058] 2. Set up each node’s network’s relative distance, speed, and direction at their initial values according to formulas (1), (2), and (3) given below:

[059] 3. Define the colliding measure as a convex function of V elrel, Distance, ?:

[060] 4. Convex Optimization (CVX Optimization) Steps:
(a) cvx begin
(b) Specify the cvx variables:
• cvx variable: K, K1,K2,K3, colliding measure parameters
• cvx variable: C(N,N), collision metric
(c) Define the objective function and constraints:
Objective Function:
Minimize

• Constraints for the collision metric:

C (i, j) = 1
• Constraints for colliding measure parameters: 0 = K,K1,K2,K3 =1
[061] 5. Normalize the collision metrics to obtain priority values between 0 and 1:
• For each node i ? ?V, calculate the priority value using the following formula:

[062] The methodology exhibits a time complexity of O(N3) and a space complexity of O(N2). These complexities are reasonable for small to moderate-sized vehicular networks but may pose challenges for larger networks.
[063] Collision avoidance warning methodology (CAWAM)
a) CAWAM operates in the Cooperative Intelligent Transportation Systems
b) (C-ITS) context to generate alerts in the event of potential collisions.
c) Real-time estimated functions F(x) and G(x) represent the speed of the ego vehicle and the distance to the vehicle ahead, respectively.
d) CAWAM employs the Recursive Least Squares (RLS) method to estimate
e) F(x) and G(x) values.
f) Alerts are triggered if F(x) increases and G(x) decreases over a specified number of incoming packets, indicating a potential collision.
g) CAWAM continuously monitors packet data and updates alert status based on changing conditions.
[064] This methodology is used in the cooperative intelligent transportation systems (C-ITS) context to generate an alert in case of a collision between the ego vehicle and the vehicle ahead. The methodology uses real-time estimated functions F(x) and G(x) to denote the speed of the ego vehicle and the distance between the ego vehicle and the vehicle ahead, respectively. The methodology generates an alert if the function F(x) increases and G(X) decreases over a certain number of incoming packets. The Recursive Least Squares (RLS)[?] method is used to estimate the values of F(x) and G(x), and the number of incoming packets determines the update rate of the functions. The main goal of the CAWA is to enhance road safety by providing early collision warning to the driver or the autonomous driving system.
[065] 1. After setting the priority values using CPAPA, take the ID and analyze the packets.
[066] 2. If for every n packet, the real-time estimated functions F(x) and G(x) are of decreasing nature, then the alert is generated.
(a) Check the decreasing nature of F(x) is updated. We see the conditions:
tn > tn-1
F(tn) > F(tn-1)
G(tn) < G(tn-1)
Where tn is the time of the packet/sequence number of the packets
for all t?t1, t2, t3......tn
(b) The functions have an update rate with n number of incoming packets is considered.
[067] The combination of CPAPAM and CAWAM methodologies leverage real-time data with advanced mathematical modeling (convex optimization and recursive least squares) to provide immediate collision prediction and avoidance, which is unique compared to slower or less adaptive existing systems.
[068] The convex optimization and recursive least squares methods in real-time within a vehicular network to predict and avoid collisions. These methods are not obvious as they require deep understanding of both vehicular networking and advanced mathematical techniques, thus providing an innovative leap over traditional collision avoidance systems.
[069] This system offers faster response times due to real-time processing. Higher precision in collision risk assessment and avoidance. Scalability across different vehicular environments, beneficial for both populated urban areas and high-speed rural settings.
[070] The input data block collects real-time data including vehicle speed, direction, and relative distance. It Feeds data to both CPAPAM and CAWAM.
[071] Collision prediction and priority value assignment methodology (CPAPAM) calculates priority values for each node using convex optimization based on collision risk metrics. It receives real-time data from the input block and sends priority values to the output block.
[072] Collision avoidance warning methodology (CAWAM) generates collision avoidance alerts using the recursive least squares method to estimate vehicle speed and distance. It receives real-time data from the input block and sends collision alerts to the output block.
[073] Output block displays the priority values and collision avoidance alerts. It receives data from both CPAPAM and CAWAM for final output.
[074] Using convex optimization techniques to dynamically calculate priority values based on real-time vehicle data provides more precise and efficient collision risk assessment compared to traditional rule-based systems.
[075] Employing the recursive least squares method for real-time estimation of vehicle speed and distance enhances the accuracy and response time of collision avoidance alerts over simpler, delay-prone calculations.
[076] The priority values and collision alerts are updated continuously based on real-time data. It allows for immediate and adaptive collision prediction and avoidance in dynamic traffic conditions.
[077] Figure 2 shows the flowchart according to the present invention. It collects real-time data including vehicle speed, direction, and relative distance. Input data is processed to calculate priority values using convex optimization and priority values are updated continuously.
[078] Vehicle speed (F(x)) and distance (G(x)) is estimated using recursive least squares and collision avoidance alerts are generated if conditions indicate a potential collision. The calculated priority values and collision avoidance alerts are displayed.
[079] Thus the system showed significant improvements in accuracy, response time, and scalability. Convex optimization and recursive least squares methods provided superior collision prediction and avoidance compared to traditional approaches. Real-time data integration ensured adaptive and immediate responses to dynamic traffic conditions.
[080] Numerous modifications and adaptations of the system of the present invention will be apparent to those skilled in the art, and thus it is intended by the appended claims to cover all such modifications and adaptations which fall within the true spirit and scope of this invention.
, Claims:WE CLAIM:
1. A system and method for real-time collision prediction and avoidance in vehicular ad hoc networks (VANETs) comprises-
a) The input data block (1) represents the data inputs required by both CPAPA and CAWA, including relative distance, speed, and direction of vehicles, as well as real-time estimated functions F(x) and G(x).
b) The collision prediction and priority value assignment methodology unit (2), which calculates priority values for each node based on collision risk metrics using convex optimization techniques.
c) The collision avoidance warning block (3) represents the collision avoidance warning methodology, which generates collision avoidance alerts based on real-time estimates of vehicle speed and distance, employing the recursive least squares method.
d) Arrows (4) indicate the flow of data and operations between the input data, CPAPA, and CAWA.
e) The output block (5) denotes the priority values assigned by CPAPA and the collision avoidance alerts generated by CAWA.
2. The method for real-time collision prediction and avoidance in vehicular ad hoc networks (VANETs), as claimed in claim 1, wherein the Collision Prediction and Priority value Assignment Methodology (CPAPAM) includes following steps:
a) CPAPAM calculates priority values for each node in the VANET based on a collision metric function.
b) Inputs to CPAPAM include relative distance, speed, and direction of vehicles within the network.
c) CPAPAM employs convex optimization techniques to minimize collision risks and assign priority values accordingly.
d) The methodology normalizes collision metrics to ensure priority values range between 0 and 1.
e) Priority values are continuously updated based on real-time vehicle data.
3. The collision avoidance warning methodology (CAWAM), as claimed in claim 2, wherein the methodology uses convex optimization to calculate priority values based on node collision risk and considers nodes’ relative distance, speed, and direction and normalizes collision metrics to assign higher priorities to nodes with higher collision risks.
4. The method for real-time collision prediction and avoidance in vehicular ad hoc networks (VANETs), as claimed in claim 1, wherein the collision avoidance warning methodology (CAWAM) includes following steps:
a) CAWAM operates in the cooperative intelligent transportation systems
b) (C-ITS) context to generate alerts in the event of potential collisions.
c) Real-time estimated functions F(x) and G(x) represent the speed of the ego vehicle and the distance to the vehicle ahead, respectively.
d) CAWAM employs the Recursive Least Squares (RLS) method to estimate
e) F(x) and G(x) values.
f) Alerts are triggered if F(x) increases and G(x) decreases over a specified number of incoming packets, indicating a potential collision.
g) CAWAM continuously monitors packet data and updates alert status based on changing conditions.
5. The collision avoidance warning methodology (CAWAM), as claimed in claim 4, wherein the methodology uses real-time estimated functions F(x) and G(x) to denote the speed of the ego vehicle and the distance between the ego vehicle and the vehicle ahead, and generates an alert if the function F(x) increases and G(X) decreases over a certain number of incoming packets,