Leveraging Augmented Reality,Semantic-Segmentation,and VANETs for Enhanced Driver’s Safety Assistance

Overtaking is a crucial maneuver in road transportation that requires a clear view of the road ahead.However,limited visibility of ahead vehicles can often make it challenging for drivers to assess the safety of overtaking maneuvers,leading to accidents and fatalities.In this paper,we consider atrous convolution,a powerful tool for explicitly adjusting the field-of-view of a filter as well as controlling the resolution of feature responses generated by Deep Convolutional Neural Networks in the context of semantic image segmentation.This article explores the potential of seeing-through vehicles as a solution to enhance overtaking safety.See-through vehicles leverage advanced technologies such as cameras,sensors,and displays to provide drivers with a real-time view of the vehicle ahead,including the areas hidden from their direct line of sight.To address the problems of safe passing and occlusion by huge vehicles,we designed a see-through vehicle system in this study,we employed a windshield display in the back car together with cameras in both cars.The server within the back car was used to segment the car,and the segmented portion of the car displayed the video from the front car.Our see-through system improves the driver’s field of vision and helps him change lanes,cross a large car that is blocking their view,and safely overtake other vehicles.Our network was trained and tested on the Cityscape dataset using semantic segmentation.This transparent technique will instruct the driver on the concealed traffic situation that the front vehicle has obscured.For our findings,we have achieved 97.1% F1-score.The article also discusses the challenges and opportunities of implementing see-through vehicles in real-world scenarios,including technical,regulatory,and user acceptance factors.

Minimum Time Overtaking Problem of Vehicle Handling Inverse Dynamics Based on Two Kinds of Safe Distances

Overtaking accidents caused by improper operations performed by a driver occur frequently. However, most stud?ies on overtaking safety have neglected research into driver control input. A novel method is proposed to obtain the driver control input during the overtaking process. Meanwhile, to improve the safety of overtaking, two types of safe distances, and the time of the overtaking are considered. Path constraints are established when considering the two types of safe distances. An optimal control model is established to solve the minimum time maneuver under multiple constraints. Using the Gauss pseudospectral method, the optimal control problem is converted into a nonlinear pro?gramming problem, which is then solved through sequential quadratic programming(SQP). In addition, the e ective?ness of the proposed method is verified based on the results of a Carsim simulation. The simulation results show that by adopting an inverse dynamics method to solve the manipulation problem of the vehicle’s minimum overtaking time, the manipulation capability of a vehicle in completing an overtaking safely within the minimum time can be obtained. This method can provide a reference for research into the active safety of manned and unmanned vehicles.

Effects of Crosswind on an Automobile Under Dynamic Conditions

This paper presents results of a series of experiments carried out on models of a car and a truck in a wind tunnel to observe the pressure coefficient variation when the truck was overtaken by the car.However,conventional wind tunnels are not equipped to measure the yaw moment and side force created due to pressure difference created as a result of the interaction of two vehicles and the cross wind.An additional attachment for the wind tunnel was designed and built to measure the yawing moment and the side force experienced by the car when the cross wind is applied.The attachment consists of a"top plate"which is independent from the wind tunnel frame and free to rotate.The car model is fixed to the"top plate"the top plate is connected to the base plate via a load measuring shaft which is connected to load cells that enable to measure the yaw moment.The pressure coefficient variation was measured by a series of pressure tapings provided by the circumference of the car.The two models were tested independently in the wind tunnel for their pressure coefficient variation.It was found that the pressure coefficient around the car varies substantially when it is overtaking the truck.It was also observed that when the cross wind is applied the yawing moment and the side force on the car increases with the air speed.The attachment was calibrated prior to the investigation and have been able to obtain consistent results.To relate the observations to real world conditions,further investigations are necessary.

Influence of roadway geometric elements on driver behavior when overtaking bicycles on rural roads

The objective of this research was to determine what influence geometric design elements of roadway may have on driver behavior during the overtaking maneuver. This was part of a larger research effort to eliminate crashes （and the resulting fatalities and injuries） between bicycles and motorized vehicles. The data collection process produced 1151 observations with approximately 40 different independent variables for each data point through direct observation, sensor logging, or derivation from other independent variables. Prior research by the authors developed a means to collect real-time field data through the use of a bicycle-mounted data collection system. The collected data was then used to model lateral clearance distance be- tween vehicles and bicycles. The developed model confirmed field observations that the lateral clearance distance provided by drivers changes with vehicle speed and oncoming vehicle presence. These observations were presented by the authors previously. The model shows that driver behavior can be adjusted by the inclusion, or exclusion, of geometric elements. Evaluating roadways （or roadway designs） based on this model will enable stakeholders to identify those roadway segments where a paved shoulder would prove an effective safety countermeasure. This research will also enable roadway designers to better identify during the design phase those roadway segments that should be constructed with a paved shoulder.

Overtaking behaviour of vehicles on undivided roads in non-lane based mixed traffic conditions

Traffic on Indian roads is highly mixed in nature with wide variations in the static and dynamic characteristics of vehicles. These vehicles do not follow strict lane discipline and occupy any available lateral position on the road space. Overtaking is one of the most complex and important manoeuvre on undivided roads where the vehicles use the opposing lane to overtake the slower vehicles with the presence of oncoming vehicles from opposite direction. They are unavoidable especially in the case of mixed traffic conditions where there is always a speed difference between the fast and slow moving vehicles. Overtaking process involves lane-changing manoeuvres, acceleration and deceleration actions and estimation of relative speed of overtaking and overtaken vehicles, and also, estimation of speed and distance of the oncoming vehicle. The main objective of the present study is to study the overtaking characteristics of vehicles on undivided roads under mixed traffic conditions. For this purpose, details of overtaking data were collected on a two-lane two-way undivided road using moving car observer method and registration plate method. The overtaking characteristics of all types of vehicles under mixed traffic conditions were observed and mathematically modelled. The data extracted and analysed were the acceleration characteristics, speeds of the overtaking vehicles, overtaking time, overtaking distances, safe opposing gap required for overtaking, flow rates, overtaking frequencies, types of overtaking strategy, and types of overtaking and overtaken vehicles. Two types of overtaking strategies were observed in the field such as flying overtaking and accelerative overtaking. Graphs were plotted between the relative speed of the overtaking and overtaken vehicles against the overtaking time and negative correlation was found between the speed differential and total overtaking time for all categories of vehicles. It was observed that the number of overtaking increases with increase in the flow rate in the on-going direction and decreases with increase in flow in the opposite direction, The results obtained from this study will be useful to understand the overtaking behaviour of vehicles in mixed and non-lane discipline traffic conditions. These parameters will be useful in the development of traffic simulations models for undivided roads and thereby for estimation of capacity. The findings from the study can also be used to estimate potential collision times which will be helpful to improve the road safety.

Autonomous Overtaking for Intelligent Vehicles Considering Social Preference Based on Hierarchical Reinforcement Learning

As intelligent vehicles usually have complex overtaking process,a safe and efficient automated overtaking system(AOS)is vital to avoid accidents caused by wrong operation of drivers.Existing AOSs rarely consider longitudinal reactions of the overtaken vehicle(OV)during overtaking.This paper proposed a novel AOS based on hierarchical reinforcement learning,where the longitudinal reaction is given by a data-driven social preference estimation.This AOS incorporates two modules that can function in different overtaking phases.The first module based on semi-Markov decision process and motion primitives is built for motion planning and control.The second module based on Markov decision process is designed to enable vehicles to make proper decisions according to the social preference of OV.Based on realistic overtaking data,the proposed AOS and its modules are verified experimentally.The results of the tests show that the proposed AOS can realize safe and effective overtaking in scenes built by realistic data,and has the ability to flexibly adjust lateral driving behavior and lane changing position when the OVs have different social preferences.

Transient aerodynamic characteristics of vans during the accelerated overtaking process

This paper studies the influence of the accelerated overtaking process on the vehicles＇ transient aerodynamic characteristics, through 3-D numerical simulations with dynamic meshes and sliding interface technique. Numerical accuracy is verified by experimental results. The aerodynamic characteristics of vehicles in the uniform overtaking process and the accelerated overtaking process are compared. It is shown that the speed variation of the overtaking van would influence the aerodynamic characteristics of the two vans, with greater influence on the overtaken van than on the overtaking van. The simulations of three different accelerated overtaking processes show that the greater the acceleration of the overtaking van, the larger the aerodynamic coefficients of the overtaken van. When the acceleration of the overtaking van increases by 1 m/s2, the maximum drag force, side force and yawing moment coefficients of the overtaken van all increase by more than 6%, to seriously affect the power performance and the stability of the vehicles. The analysis of the pressure fields under different accelerated conditions reveals the cause of variations of the aerodynamic characteristics of vehicles.

Simulation and Analysis on Overtaking Safety Assistance System Based on Vehicle-to-Vehicle Communication

Because of the complexity and variability of an intelligent vehicle’s driving environment,it is difficult for the application of the vehicular sensors to meet the needs of the surrounding environment information entirely.Vehicle-to-vehicle(V2V)communication technology is used by target vehicles to exchange information,and obtain the driving condition and driving intention of the front driver.To obtain environmental information outside the range of vehicular sensors in advance,in this paper,a vehicle overtaking assistance system is proposed based on V2V communication.The data,including the speed,position,direction angle and steering angle obtained using V2V communication,were preliminarily processed.Then,combined with an overtaking safety distance model,the vehicle parameters,driver’s driving intention and vehicle status information were entered into an overtaking security assistance system to determine the overtaking conditions.Fuzzy theory was used to control the parameters of the overtaking safety distance model.Finally,the overtaking safety assistance system was established and the proposed algorithm was tested using PreScan/MATLAB cooperative simulation software.The results showed that the proposed overtaking safety algorithm effectively provided a warning according to environmental change and the driver’s intention,which assisted the driver to overtake and avoid the occurrence of accidents,which improved the safety performance of the vehicle.

A fully nonlinear and weakly dispersive water wave model for simulating the propagation,interaction,and transformation of solitary waves

This paper presents the development of a theoretical model of fully nonlinear and weakly dispersive(FNWD)waves and numerical techniques for simulating the propagation,interaction,and transformation of solitary waves.Using the standard expansion method and without the limit of small nonlinear parameter defined as the ratio of the wave height versus water depth,a set of model equations describing the FNWD waves in a domain of moderately varying bottom topography are formulated.Exact solitary wave solutions satisfying the FNWD equations are also derived.Numerically,a time-accurate and stabilized finite-element code to solve the governing equations is developed for wave simulations.The solitary wave solutions of FNWD,weakly nonlinear and weakly dispersive(WNWD),and Laplace equations based models in terms of wave profile and phase speed are compared to examine their related features and differences.Investigations on the overtaking collision of two unidirectional solitary waves of different amplitudes,i.e.,ax and a2 where a1>a2,are carried out using both the FNWD and WNWD water wave models.Selected cases by running the FNWD and WNWD models are performed to identify the critical values of a1/a2 for forming a flattened merging wave peak,which is the condition used to determine if the stronger wave is to pass through the weaker one or both waves are to remain separated during the encountering process.It is interesting to note the critical values of a1/a2 obtained from the FNWD and WNWD models are found to be different and greater than the value of 3 proposed by Wu through the theoretical analysis of the Korteweg-de Vries(KdV)equations.Finally,the phenomena of wave splitting and nonlinear focusing of a solitary wave propagating over a three-dimensional semicircular shoal are simulated.The results obtained from both the FNWD and WNWD models showing the fission process of separating a main solitary wave into multiple waves of decreasing amplitudes are presented,compared,and discussed.

Simulation of the overtaking maneuver between two ships using the non-linear maneuvering model

The overtaking maneuver of two Esso Bernicia 1.9×10^5 DWT tankers was investigated using the system-based maneuvering method. The Brix model was incorporated to account for the overtaking interactions. Forces/moment acting on the ship hulls and maneuvering motions were analyzed for both ships during the overtaking. The influences of the speed ratio and the passing distance were specially studied. Two fitting formulas for the minimum distance during the overtaking were finally established as functions of the two factors respectively. They can be used to predict the minimum distance during an overtaking maneuver under similar conditions to avoid marine accidents.

An Autonomous Overtaking Maneuver Based on Relative Position Information

Reliable and efficient overtaking maneuvers are important and challenging for autonomous vehicles.Higher precision position information is thus required,rather than that traditional global navigation satellite systems can provide.In this paper,we try to perform reliable autonomous overtaking controls of vehicles,mainly based on the“relative”position information,including the distance,angle and velocity between vehicles,which can be achieved by on-board radars.To reduce the complexity of maneuvers,a fuzzy inference system is applied to analyze the driving behavior of the preceding vehicle based on the obtained consecutive relative position information.An output of“safe”or“dangerous”will be sent to the decision part based on reinforcement learning frameworks.Various overtaking maneuvers including“conservative”and“aggressive”can be obtained accordingly.Furthermore,we propose another overtaking strategy that vehicles can share their maneuver information during overtaking process via wireless links.Numeric results validate our analysis,and can provide meaningful performance benchmarks for practical system implementations.