Vehicle Path Prediction Research Articles

Self‐adaptive search optimization‐based vehicle path prediction and traffic light controller in vehicular ad hoc network

SummaryIn recent days, VANET is considered as the main hopeful equipment in the system of transportation since traffic congestion arises regularly and thus occurred road accidents very easily. Besides, the network traffic is increasing due to the huge quantity of information generated in region of urban. Therefore, one of the primary challenges faced by Intelligent Transportation System (ITS) is ensuring the accurate transmission of information in both vehicle‐to‐vehicle (V2V) and vehicle‐to‐road (V2R) sensing unit communications. So, here, an adaptive routing controller (ARC) is utilized to improve the data transmission between vehicle and road sensing unit (RSU) and to minimize the traffic density in VANET, and the self‐adaptive search optimization is developed in VANET clustering to prioritize vehicles in the lane, in which the multi‐objective function is intended depending on the energy of the node, acceleration, jitter, priority, velocity, and trust factors. The traffic light control is done to facilitate the effective communication in the network. Thus, the proposed technique is calculated in terms of throughput, jitter, quadratic mean of acceleration (QMA), and spatially distributed travel time (SDTT), which acquired the values of 0.53 s, 36.07 kmph, 3.472 s, and 43.572%, respectively, while using 50 vehicles at 50 s.

A Novel Multimodal Vehicle Path Prediction Method Based on Temporal Convolutional Networks

Accurate and reliable prediction of future motions of the nearby agents and effective environment understanding will contribute to high-quality and meticulous path planning for the automated vehicles under uncertainty and guarantee traffic safety for future real-world deployments. This task becomes more challenging in highly dynamic and complex scenarios such as unsignalized intersections where no lights exist to control vehicles behavior, or there are not multiple lines for the vehicles to anticipate drivers’ future intentions based on the lane in which they are driving. In this study, we introduce a novel deep learning-based methodology to anticipate vehicles path at unsignalized intersections. The method provides multimodal outputs to take into account the inherited uncertainty and multimodality nature of vehicles behavior. Our proposed model works based on dilated convolutional networks in combination with a mixture density layer. We then cluster various existing mixes into possible paths that are ranked based on probability. We assess the performance and generalization capability of our vehicle path prediction model using several metrics over a large naturalistic dataset containing more than 23800 vehicle trajectories. The obtained results reveal the higher performance of our path prediction approach compared with several baselines and benchmarks.

STAG: A novel interaction-aware path prediction method based on Spatio-Temporal Attention Graphs for connected automated vehicles

Understanding social interactions between a vehicle and its surrounding agents enables effective path prediction, which is critical for the motion planning and safe navigation of automated vehicles. Several existing studies adopt recurrent neural networks which are powerful in modeling temporal interactions. Nevertheless, these models fail to deal with the complex spatio-temporal correlations among the traffic participants, especially for long-term predictions. On the other hand, recent spatio-temporal approaches based on Graph Neural Networks (GNN) have demonstrated considerable potential in modeling the social spatial and temporal interactions between the neighboring agents. These approaches, however, consider the influence of distinct agents on one another within a scene either fixed or symmetric, which may not be true in real-world scenarios. In this work, we introduce a novel graph-based system for vehicle path prediction, namely Spatio-Temporal Attention Graph (STAG). More specifically, STAG explicitly activates social interaction modeling and relational reasoning with a directed graph representation while considering spatial inter-agent correlations and motion tendency aspects. We assess the performance of STAG in three different driving scenarios, including highly structured highways, complex roundabouts, and highly dynamic unsignalized intersections. The results indicate that STAG outperforms several benchmark methods in terms of path prediction performance.

RNN-Based Path Prediction of Obstacle Vehicles With Deep Ensemble

In this paper, a new approach for obstacle vehicle path prediction, which is important for advanced driver assistance systems (ADAS) and autonomous vehicles, is proposed based on a deep neural network. In order to analyze sequential sensor data, a recurrent neural network (RNN) is used and the input data for RNN is drawn from three sensors: LIDAR, camera and GPS. These sensor data are obtained experimentally with real vehicles. In addition, deep ensemble is used for robustness of the estimation and acquisition of the uncertainty. The predicted path of the proposed method is continuous and it predicts both short-term and long-term path with a single algorithm. The size of the network model is small, but it shows good performance in predicting future trajectory of obstacle vehicles.

Vehicle Path Prediction Using Yaw Acceleration for Adaptive Cruise Control

In this paper, we propose a vehicle path prediction employing yaw acceleration for adaptive cruise control (ACC). First, a path prediction method employing yaw acceleration is proposed to improve the path prediction performance of ego vehicles. In the proposed method, the vehicle path is predicted by using a clothoidal cubic polynomial curve model, and for this purpose, the curvature rate, yaw rate, and longitudinal velocity are required. The curvature rate can be mathematically obtained by differentiating the yaw rate without a camera sensor. To obtain the yaw acceleration from the noisy measured yaw rate, we derive the state-space model from the steer wheel angle for the yaw rate. Then, the KF is designed by using the state-space model to estimate the yaw acceleration. Second, a multirate longitudinal control method is proposed to improve the longitudinal control performance. The multirate KF employs the constant acceleration model in order to estimate the relative distance and velocity of the target vehicle at a faster sampling rate. Then, the desired acceleration is achieved to maintain a safe headway distance or velocity by means of the longitudinal controller. Consequently, the whole ACC system operates with a faster sampling rate so that multirate control scheme reduces ripples in both the relative longitudinal distance and the desired acceleration. The performance of the proposed method was evaluated via simulations and experiments.

Intelligent braking system for stability enhancement of vehicle braking, using fuzzy logic controllers

In this research, a novel structure is proposed for vehicle braking stability improvement encountering critical driving conditions such as braking on slippery or μ-split surface. For these purposes, at first, a trained neuro-fuzzy estimator is used for vehicle path prediction according to vehicle speed, applied steering angle and also their changes. Then, slip of each wheel will be controlled by an anti-slip fuzzy controller which has been designed for each wheel. Also, another fuzzy controller has been designed so as to control the yaw angle concerning the yaw error. Then, the difference between the left and the right wheels’ braking torques is used by the ASBS controller in order to decrease the yaw error. Considering a model of Three-Degree-of-Freedom for chassis modelling and One-Degree-of-Freedom Dugoff’s tyre model for each wheel, a series of Matlab/Simulink simulation results will be presented in order to validate the effectiveness of the proposed controller.

INTELLIGENT BRAKING SYSTEM USING FUZZY LOGIC AND SLIDING MODE CONTROLLER

In this paper, an antilock–antiskid braking system controller, which has been designed for stability enhancement of vehicles during braking and turning, is presented. Using available signals, a novel structure is proposed for vehicle stability improvement for critical driving conditions such as braking on slippery or µ-split road surfaces. In conventional vehicles, undesired lane changes may occur due to equal dispatch of braking torques to all wheels simultaneously. Also, intensive pressure on brake pedal can bring about wheel lockup which results in vehicle instability and undesired lane changes. Antilock braking system (ABS) along with Antiskid braking system (ASBS) can serve as a driver-assistance system in vehicle path correction facing critical driving conditions during braking and turning round on different road surfaces. For these purposes, at first, a trained Neuro-Fuzzy estimator is used for vehicle path prediction according to vehicle’s speed, applied steering angle and their changes. Then, slip of each wheel will be controlled by an antilock fuzzy controller which has been designed for each wheel. Also, a sliding mode controller has been designed so as to control the yaw angle considering the yaw error, where the yaw error is resulted from the difference of the Neuro-Fuzzy estimator’s output and actual yaw angle. Then, the difference of the left and the right wheels’ braking torques are used by the sliding mode controller in order to reduce the yaw error. Considering a model of three-degreeof-freedom for chassis and one-degree-of-freedom Dugoff’s tire model for each wheel, a series of Matlab/Simulink simulation results will be presented to validate the effectiveness of the proposed controller. Finally, a comparison will be made between the proposed method and one of the recent control systems which shows the superiority of its performance.

Vehicle dynamics and external disturbance estimation for vehicle path prediction

This paper addresses the onboard prediction of a motor vehicle's path to help enable a variety of emerging functions in autonomous vehicle control and active safety systems. It is shown in simulation that good accuracy of path prediction is achieved using numerical integration of a linearized two degree of freedom vehicle handling model. To improve performance, a steady-state Kalman filter is developed to estimate the vehicle's lateral velocity and the magnitudes of external disturbances acting on the vehicle, specifically the lateral force and the yaw moment disturbances. A comparison is made between three models of external disturbance time variation; a piecewise-constant-in-time model is found to be sufficient. Finally, an algorithm is proposed to characterize path prediction uncertainty using a statistical characterization of the measurement and modeling errors. Simulation suggests that these algorithms may provide a useful suite of path prediction tools for a variety of applications.