Driver Actions Research Articles

Design and Evaluation of a Human-in-the-Loop Connected Cruise Control

The performance of automated vehicle can be greatly improved by enabling vehicle-to-vehicle communication. An example of such applications is Cooperative Adaptive Cruise Control (CACC) along the highway. Although the adoption rate of vehicular connectivity is predicted to grow rapidly, CACC can only benefit vehicles that are both connected and automated. To take a full advantage of vehicular connectivity, a human-in-the-loop Connected Cruise Control (hCCC) algorithm is developed for human-driven connected vehicle. In hCCC, the human driver remains engaged in the longitudinal control of the vehicle, and hCCC controller applies additional acceleration/deceleration on top of human actions according to the received status of preceding vehicle. By allowing coexistence of the automatic control and driver's actions in a beneficial way, hCCC helps the human driver stabilize the vehicle more efficiently and safely. The proposed hCCC inherits the feedback-feedforward control structure and velocity-dependent spacing policy from the typical CACC systems. String stability analysis shows that hCCC can offer broader string-stable ranges of human parameters than human driving alone or the existing acceleration-based Connected Cruise Control (CCC), indicating a better capability to mitigate traffic disturbance with the uncertain human behaviors. The desirable properties of hCCC were validated in driving simulator experiments, which showed that hCCC could reduce 36.8% acceleration, 31.2% time-gap fluctuation, 81.2% exposure time to unsafe driving situations, and 15.8% fuel consumption from those of human driving alone. In addition, two derivative designs of hCCC are proposed and proven effective, further lowering down the practice threshold of hCCC.

The Multimodal Driver Monitoring Database: A Naturalistic Corpus to Study Driver Attention

A smart vehicle should be able to monitor the actions and behaviors of the human driver to provide critical warnings or intervene when necessary. Recent advancements in deep learning and computer vision have shown great promise in monitoring human behaviors and activities. While these algorithms work well in a controlled environment, naturalistic driving conditions add new challenges such as illumination variations, occlusions and extreme head poses. A vast amount of in-domain data is required to train models that provide high performance in predicting driving related tasks to effectively monitor driver actions and behaviors. Toward building the required infrastructure, this paper presents the multimodal driver monitoring (MDM) dataset, which was collected with 59 subjects that were recorded performing various tasks. We use the Fi- Cap device that continuously tracks the head movement of the driver using fiducial markers, providing frame-based annotations to train head pose algorithms in naturalistic driving conditions. We ask the driver to look at predetermined gaze locations to obtain accurate correlation between the driver's facial image and visual attention. We also collect data when the driver performs common secondary activities such as navigation using a smart phone and operating the in-car infotainment system. All of the driver's activities are recorded with high definition RGB cameras and time-of-flight depth camera. We also record the controller area network-bus (CAN-Bus), extracting important information. These high quality recordings serve as the ideal resource to train various efficient algorithms for monitoring the driver, providing further advancements in the field of in-vehicle safety systems.

Vehicle drivetrain and fuzzy controller optimization using a planar dynamics simulation based on a real-world driving cycle

The plug-in hybrid electric vehicle (PHEV) stands out among hybrid vehicles due to the ability to recharge the battery from an external source and improve vehicle performance and efficiency by increasing fuel economy and reducing emissions. Due to the use of two power sources, a power management strategy must be determined to ensure the operation of both systems achieves higher global efficiency. Therefore, this paper presents three fuzzy controllers. Two controllers act hierarchically to perform the power management, and the third controller is responsible for performing the gear shift. A multi-objective optimization based on the particle swarm algorithm is proposed to reduce the driver's action on the steering angle, minimize the mass of electrical components (electric motors and battery), the equivalent fuel consumption, and exhaust emissions. To do this, the electric motor torque curve, battery capacity and voltage, fuzzy controller parameters are optimized, and the best gearbox and mechanical differential configuration are determined. The best trade-off solution, when compared to the conventional vehicle, was able to improve the handling performance with a 71.9% reduction in the driver steering action, decrease fuel consumption by 18.44%, and reduce pollutant emissions by 69.34% of CO, 20.33% of HC, and 22.11% of NOx.

Predicting the Degree of Distracted Driving Based on fNIRS Functional Connectivity: A Pilot Study.

Distracted driving is one of the main causes of traffic accidents. By predicting the attentional state of drivers, it is possible to prevent distractions and promote safe driving. In this study, we developed a model that could predict the degree of distracted driving based on brain activity. Changes in oxyhemoglobin concentrations were measured in drivers while driving a real car using functional near-infrared spectroscopy (fNIRS). A regression model was constructed for each participant using functional connectivity as an explanatory variable and brake reaction time to random beeps while driving as an objective variable. As a result, we were able to construct a prediction model with the mean absolute error of 5.58 × 102 ms for the BRT of the 12 participants. Furthermore, the regression model with the highest prediction accuracy for each participant was analyzed to gain a better understanding of the neural basis of distracted driving. The 11 of 12 models that showed significant accuracy were classified into five clusters by hierarchical clustering based on their functional connectivity edges used in each cluster. The results showed that the combinations of the dorsal attention network (DAN)-sensory-motor network (SMN) and DAN-ventral attention network (VAN) connections were common in all clusters and that these networks were essential to predict the degree of distraction in complex multitask driving. They also confirmed the existence of multiple types of prediction models with different within- and between-network connectivity patterns. These results indicate that it is possible to predict the degree of distracted driving based on the driver's brain activity during actual driving. These results are expected to contribute to the development of safe driving systems and elucidate the neural basis of distracted driving.

Multinomial Logistic Regression Modeling of Motorcycle Crash Severities and Contributing Factors in Wyoming

Motorcycle riders and passengers are much more likely to be killed or severely injured in a crash, and on average about 15% of all traffic fatalities include motorcyclists. Between 2008 and 2019, the average motorcycle crash frequency in Wyoming was 286 crashes/year, 17 of those being fatal. This paper assesses injury severity of motorcycle-related crashes in Wyoming using 12 years of motorcycle crash data and applying multinomial logistic regression modeling to determine the odds ratios for injury severity. Four models were developed and analyzed, based on the setting and the number of vehicles involved. The most common factors affecting injury severity include vehicle maneuver, driver action, junction relation, alcohol, animal and speed involvement, and helmet use. The vicinity of intersections significantly increases the odds of injury crashes in urban areas, and in rural areas with multi-vehicle involvement. Certain vehicle maneuvers are also associated with a more severe crash outcome.

IMPROVING THE TRAFFIC SIGNS ACCURACY DETECTION WITH DIGITAL CALIBRATION OF WIDE-ANGLE CAMERA

Problems. The analysis of statistical data shows that from year to year there are more and more vehicles on public roads, as result, the number of accidents increases and the risk for each road user increases. To ensure road safety, automakers are implementing driver assistance systems and cars with autonomous control elements. Such systems are capable of analyzing the environment and in the dangerous situations can correct the driver's actions or taking control of the vehicle. Typically, such systems work with the television systems help, one of the requirements for which is a wide angle of view. However, the main disadvantage of wide-angle systems is the large distortion, ie the geometric distortion of the similarity between the object and its image.
 Objective. Develop an approach and algorithm for calibration of wide-angle cameras for autonomous driving.
 Methods. Analysis and modeling of television car control system and its calibration.
 Results. A comparison of wide-angle and conventional television systems was performed, each was calibrated and the analysis of the location of road signs of each of the systems was performed.
 Conclusions. It has been shown that for the purposes of driving a car, it is possible to use wide-angle television systems with their subsequent calibration to eliminate geometric distortion between the object and its image. It is possible to use wide-angle television car control systems with a large value of distortion after its calibration. Studies have shown that the IoU metric on a wide-angle camera is almost the same as on a conventional camera, and the road sign was found even at the edge of the field of view.
 The application of the developed method of calibration of wide-angle cameras in experiments preserves the geometric similarity between the object and its image at a much larger angle of view in such a camera. Image processing time is insignificant, because the current level of computer power allows you to process such operations in real time.

Understanding the influence of Iranian farmers’ climate change beliefs on their adaptation strategies and mitigation intentions

ABSTRACT Agriculture is one of the most vulnerable sectors to climate change, therefore it is necessary to understand farmer perceptions of – and adaptation to – climate change. As part of a unique farm adaptation study in Iran, we used structural equation modelling to analyze 400 on-farm survey responses from Khuzestan Province farmers, to understand how their different types of climate change beliefs influenced their mitigation and adaptation behaviour. The structural equation modelling results show that: (1) social trust and responsibility directly influenced adaptation and mitigation; (2) age, education, and family size were not significantly associated with the adaptation and mitigation indexes; and (3) believing in overall climate change was not the actual driver of farmer adaptation (or mitigation) behaviour, and that overall beliefs were mediated through individual farm and local experience. Beliefs in farm and local climate change impacts had the largest, positive, and significant influence on farm adaptation. Once mediated, overall farmer climate change beliefs played a negative role influencing total farm adaptation behaviour, highlighting the need for government and agencies to target communication around the risk impacts of climate change, particularly at the local and regional levels.

Modelling of Pedals Pressing Mechanisms for Low Speed Driving in Road Traffic Delay

Road traffic delay is imminent in big cities around the world including Kuala Lumpur, due to the increase in vehicle ownership, weather condition, road blocks by authorities and ongoing infrastructure projects at particular point of the road. Previous literature shows that the average time spent by drivers in driving under road traffic delay is about 30-53 minutes per day. Within the delay, regular pressings of the brake and accelerator may lead to leg and body fatigue, hence may give detrimental effect on the driver in the long run. This paper aims to mimic the pedal pressing and releasing mechanism in replacing the leg activities during road traffic delay, such that the driver will no longer need to recurrently pressing both pedals, thus reduce the fatigue. In replicating the pressing mechanism, a linear electromechanical actuator is designed for this purpose due to its high force density as well as good response time. The mathematical model of the actuator is first developed and integrated with car dynamic model in Simscape model. All parameters used for simulation purposes were taken from an actual electromechanical actuator used for the similar application. Simulation performance evaluation was conducted on a virtual environment for the integrated models and the response of the models were investigated. As a result, the actuator model is capable to produce the desired output force that is similar with human leg during pressing activity, with about 200 N force. The model was coupled with tuned PID controller, and the actuator was also able to produce an acceptable time response, where the vehicle is brought to a complete stop within less than 1 second, which is closed to the actual driver’s leg activities during the pedal pressing in a road traffic delay.

Evaluation of 1D and 2D Deep Convolutional Neural Networks for Driving Event Recognition.

Driving event detection and driver behavior recognition have been widely explored for many purposes, including detecting distractions, classifying driver actions, detecting kidnappings, pricing vehicle insurance, evaluating eco-driving, and managing shared and leased vehicles. Some systems can recognize the main driving events (e.g., accelerating, braking, and turning) by using in-vehicle devices, such as inertial measurement unit (IMU) sensors. In general, feature extraction is a commonly used technique to obtain robust and meaningful information from the sensor signals to guarantee the effectiveness of the subsequent classification algorithm. However, a general assessment of deep neural networks merits further investigation, particularly regarding end-to-end models based on Convolutional Neural Networks (CNNs), which combine two components, namely feature extraction and the classification parts. This paper primarily explores supervised deep-learning models based on 1D and 2D CNNs to classify driving events from the signals of linear acceleration and angular velocity obtained with the IMU sensors of a smartphone placed in the instrument panel of the vehicle. Aggressive and non-aggressive behaviors can be recognized by monitoring driving events, such as accelerating, braking, lane changing, and turning. The experimental results obtained are promising since the best classification model achieved accuracy values of up to 82.40%, and macro- and micro-average F1 scores, respectively, equal to 75.36% and 82.40%, thus, demonstrating high performance in the classification of driving events.

Modeling factors to various drivers' actions due to vehicle malfunctions, the integrated latent class and latent variable model

It is known that how individuals perceive the situation result in choices. However, it is unknown, how drivers process the information of vehicle malfunctions based on the situation at hand. In this study, we only considered drivers' actions in cases where vehicles had some forms of malfunction at the time of the crash. Also due to endogeneity bias, or aggregation of the error term, we did not include the vehicle malfunction directly in our model. In addition, we used the latent variable (LV) for class allocation. Although the latent class model considering latent variable (LCLV) is not comparable with other considered models, by means of log likelihood, the results expanded our understanding regarding the indirect impact of an indicator of vehicle malfunction by means of latent variable on drivers' actions. Due to inclusion of latent variables in both, the indicator and the choice models, the relationship between those two models could be revealed based on the impacts of the latent variables. It was found that, for instance, the latent variable, which decreases the likelihood of driving vehicles with brake defect, at the same time increases taking driver action of speeding. Or female drivers are less likely to be involved in speeding driving actions before crashes due to various vehicles malfunctions. Driving under the influence of drug or alcohol or being under emotional conditions were some of the factors impacting various drivers actions. This is the earliest study implemented the technique in the context of traffic safety.

A pose-aware dynamic weighting model using feature integration for driver action recognition

Traffic accidents caused by distracted driving are on the rise, posing a serious threat to the safety of people’s lives and property. Recognition and early warning of the driver’s actions is particularly important. Considering the differences in local details of driver actions, we use the keypoint information of drivers that reflects the category differences. Specifically, we explicitly model keypoints features and propose a pose-aware driver action recognition model. We design a pose-based feature fusion module incorporating the attention mechanism, to fuse the global features and keypoint features of different scales in driving images. In addition, we propose an input-dependent weighting module to enhance the discrimination of the fused features. We use dynamic convolution to apply channel attention to convolution weights. According to the input driving image, the corresponding weights are adaptively generated for multiple convolution kernels, and the final weighted summation is carried out. This is a soft gate scheme, which opens a new perspective for driver action recognition. The proposed model achieves 90.7% and 95.3% accuracy on StateFarm dataset and SEU-Driving dataset, which are improved by 1.4% and 3.1% respectively compared to SOTA (MSA-CNN).

Building a mathematics model for lane‐change technology of autonomous vehicles

In the process of autonomous vehicle motion planning and to create comfort for vehicle occupants, factors that must be considered are the vehicle's safety features and the road's slipperiness and smoothness. In this paper, we build a mathematical model based on the combination of a genetic algorithm and a neural network to offer lane-change solutions of autonomous vehicles, focusing on human vehicle control skills. Traditional moving planning methods often use vehicle kinematic and dynamic constraints when creating lane-change trajectories for autonomous vehicles. When comparing this generated trajectory with a man-generated moving trajectory, however, there is in fact a significant difference. Therefore, to draw the optimal factors from the actual driver's lane-change operations, the solution in this paper builds the training data set for the moving planning process with lane change operation by humans with optimal elements. The simulation results are performed in a MATLAB simulation environment to demonstrate that the proposed solution operates effectively with optimal points such as operator maneuvers and improved comfort for passengers as well as creating a smooth and slippery lane-change trajectory.

Real-World Crash Circumstances Among Newly Licensed Adolescent Drivers With and Without Attention-Deficit/Hyperactivity Disorder

PurposeAdolescents with attention-deficit/hyperactivity disorder (ADHD) have 30%–40% higher crash rates. However, we still do not understand which factors underlie heightened crash risk and if crash circumstances differ for drivers with ADHD. We compared prevalences of crash responsibility, driver actions, and crash types among adolescent and young adult drivers with and without ADHD who crashed within 48 months of licensure. MethodsIn this exploratory retrospective cohort study, we identified patients of Children's Hospital of Philadelphia's (CHOP) New Jersey (NJ) primary care locations who were born between 1987 and 2000, NJ residents, had their last CHOP visit ≥ age 12 years, and acquired a driver's license. We linked CHOP electronic health records to NJ's licensing and crash databases. ADHD diagnosis was based on International Classification of Diseases, Ninth Revision, Clinical Modification/International Classification of Diseases, Tenth Revision, Clinical Modification codes. Prevalence ratios were estimated using generalized estimating equation log-binomial regression. ResultsWe identified 934 drivers with ADHD in 1,308 crashes and 5,158 drivers without ADHD in 6,676 crashes. Within 48 months postlicensure, drivers with ADHD were more likely to be at fault for their crash (prevalence ratio: 1.09 [1.05–1.14]) and noted as inattentive (1.15 [1.07–1.23]). With the exception that drivers with ADHD were less likely to crash while making a left/U-turn, we did not find substantial differences in crash types by diagnosis. Analyses also suggest females with ADHD may have a higher risk of colliding with a nonmotor vehicle and crashing due to unsafe speed than females without ADHD. DiscussionThe results suggest crash circumstances do not widely differ for drivers with and without ADHD but highlight several factors that may be particularly challenging for young drivers with ADHD.

Neural network model for recognition and classification of types of interactions in road traffic

The article presents neural network for recognition of driving strategies based on interactions between drivers in road traffic. It analyzes the architecture of the model implemented as a self-organizing map (SOM), consisting of a group of neural networks based on radial basis functions (RBF). It is a training model grounded in the biological foundations of artificial neural networks, in which the training set should consist exclusively of input vectors; wherein the network training algorithm adjusts itself the network's weights to obtain consistent output vectors (i.e. to make presenting sufficiently close input vectors result into the same outputs). The article presents the results of using a new generation of the neural network developed by us, which includes an adaptive learning algorithm to reduce the effect of re-training (overfitting) and false recognition, as well as to improve the determination of the boundaries between clusters. The aim of the research is to outline architecture and structure of the neural network model that allows recognizing strategical characteristics of driving and can identify strategies of interactions between vehicles (their drivers) in road traffic as well as identify behavioral patterns This paper considers driving strategies that characterize the interaction of dyads of vehicles (drivers) moving in road traffic. The research results show that the SOM RBF neural networks can recognize and classify types of interactions in road traffic based on modelling of the analysis of vehicle movement trajectories. Experimental results demonstrate the neural networks architecture and networks learning involving 400 iterations of streaming the training data representing 500 possible simulated interaction situations. This paper presents a novel neural network model for recognition of drivers' behaviour patterns and for classification of driving strategies into five general classes: (1) competition strategy, (2) contest strategy, (3) evasion strategy, (4) compromise strategy, and (5) active confrontation strategy. This neural network has demonstrated a high rate of recognition and concise clusterization of similar driving strategies. The key contribution of this paper: it proposes a neural network model based on Kohonen's Self-Organizing Map (SOM) for detecting drivers' behaviours from vehicle movement patterns - driving strategies - instead of monitoring driver's specific activities.

Some aspects of studying the interaction of pedestrians and drivers of vehicles

The object of the research in this article is the social relations developing in the field of pedestrian safety. The subject of the study is statistical data on the state of road safety, methods and results of studies conducted to assess the safety of pedestrian traffic at regulated and unregulated pedestrian crossings. The authors study the results of recent studies of the influence of the nature of pedestrian traffic on the actions of drivers, since these variables reflect behavior ultimately associated with conflicts between vehicles and pedestrians. It was concluded that there is a strong correlation between the behavior of pedestrians and the behavior of the driver. As the main factors influencing the reduction of the probability of pedestrians passing by drivers, the high speed of traffic flow and the low possible speed of deceleration of vehicles in the flow are indicated. The novelty of the study lies in the fact that the analysis of all the latest methods for assessing pedestrian safety has allowed us to identify a number of problems that need to be solved, first of all, to reduce pedestrian deaths on the roads. In each of the studies presented in the article, different data, variables and methods were used, there was no general approach, which currently does not allow us to clearly identify risk factors and, consequently, common methods that need to be developed and implemented.

Social life cycle assessment of green methanol and benchmarking against conventional fossil methanol

Green methanol could play a major role in decarbonising both the chemical and energy sectors. While techno-economic and environmental studies on green methanol following a life-cycle perspective are available, its social implications from a supply-chain standpoint remain largely unexplored. In order to fill this gap, this work presents the first social life cycle assessment of green methanol produced with CO2 directly captured from the air and hydrogen from wind power electrolysis. When compared to conventional methanol from natural gas, the results suggest an unfavourable performance of green methanol under negative social indicators (forced labour, women in the sectoral labour force, health expenditure, social responsibility promotion, and fair salary) due to the increased supply-chain complexity of the green system. In contrast, green methanol would outperform its conventional counterpart in terms of sectoral contribution to economic development, a positive social indicator which would benefit from the increase in working hours. Besides future consideration of a higher number of positive indicators and potential improvements in country- and sector-specific risk levels towards high-quality social and working conditions, an enhanced social life-cycle performance of green methanol requires technical improvements to reduce the high demand for energy inputs and equipment across its supply chain. Acknowledging decarbonisation as the actual driver of green methanol deployment, future social studies are suggested to focus on the comparison of renewable alternatives for its production and the effect of social regulations.

A Framework for Building Comprehensive Driver Profiles

Conventional approaches to modelling driver risk have incorporated measures such as driver gender, age, place of residence, vehicle model, and annual miles driven. However, in the last decade, research has shown that assessing a driver’s crash risk based on these variables does not go far enough—especially as advanced technology changes today’s vehicles, as well as the role and behavior of the driver. There is growing recognition that actual driver usage patterns and driving behavior, when it can be properly captured in modelling risk, offers higher accuracy and more individually tailored projections. However, several challenges make this difficult. These challenges include accessing the right types of data, dealing with high-dimensional data, and identifying the underlying structure of the variance in driving behavior. There is also the challenge of how to identify key variables for detecting and predicting risk, and how to combine them in predictive algorithms. This paper proposes a systematic feature extraction and selection framework for building Comprehensive Driver Profiles that serves as a foundation for driver behavior analysis and building whole driver profiles. Features are extracted from raw data using statistical feature extraction techniques, and a hybrid feature selection algorithm is used to select the best driver profile feature set based on outcomes of interest such as crash risk. It can give rise to individualized detection and prediction of risk, and can also be used to identify types of drivers who exhibit similar patterns of driving and vehicle/technology usage. The developed framework is applied to a naturalistic driving dataset—NEST, derived from the larger SHRP2 naturalistic driving study to illustrate the types of information about driver behavior that can be harnessed—as well as some of the important applications that can be derived from it.

ПОВЫШЕНИЕ БЕЗОПАСНОСТИ ТРАНСПОРТНОЙ РАБОТЫ АВТОМОБИЛЯ НА ОСНОВЕ КОНТРОЛЯ ПСИХОЭМОЦИОНАЛЬНОГО СОСТОЯНИЯ ВОДИТЕЛЯ

A method for assessing the psychoemotional state of a car driver is proposed based on monitoring his actions according to the following six indicators: longitudinal, lateral and vertical accelerations, exceeding the speed limit, smoking the driver and his conversations distracting from tracking traffic safety. As an evaluation parameter, the coefficient of psychoemotional reliability of the driver is proposed, which provides information about the output of controlled indicators for the constants of the established values when the driver's actions cause a threat to traffic safety. A constructive scheme of the device is proposed that collects and processes information on the speci-fied indicators of the driver's actions with the issuance of a warning signal to him or the controlling entity about the occurrence of danger when the driver continues to work in the previous mode.

Дослідження фактору швидкості та методик визначення швидкості при проведенні автотехнічної експертизи

The publication notes that the rapid motorization of Ukraine has naturally led to the need to ensure road safety at a new modern level. It has been established that one of the effective measures to improve the road safety system is speed control. This, in turn, involves appropriate regulation of the permitted speeds. Determining the speed of vehicles helps to establish a causal link between the driver's actions and the accident. It is concluded that the most significant cause of road accidents with serious consequences and deaths is the wrong choice of speed. The following issues require further, more in-depth study: methodological aspects of determining the speed of movement of ATZ, which are based on the use of the laws of conservation of the amount of movement; construction of a mathematical model of the interaction of a car wheel with a boundary obstacle; statistical studies on the formation and updating of the database on stiffness and damping parameters of modern suspensions and the coefficient of adhesion; comparative assessment of the probability of traditional and improved methods of calculating vehicle speeds before a collision in a road accident; development of proposals for instrumental provision of a mobile laboratory for the reconstruction of road accidents at the scene of the incident.

Planning for Autonomous Driving via Interaction-Aware Probabilistic Action Policies

Devising planning algorithms for autonomous driving is non-trivial due to the presence of complex and uncertain interaction dynamics between road users. In this paper, we introduce a planning framework encompassing multiple action policies that are learned jointly from episodes of human-human interactions in naturalistic driving. The policy model is composed of encoder-decoder recurrent neural networks for modeling the sequential nature of interactions and mixture density networks for characterizing the probability distributions over driver actions. The model is used to simultaneously generate a finite set of context-dependent candidate plans for an autonomous agent and to anticipate the probable future plans of human drivers. This is followed by an evaluation stage to select the agent plan with the highest expected utility for execution. Our approach leverages rapid sampling of action distributions in parallel on a graphic processing unit, offering fast computation even when modeling the interactions among multiple vehicles and over several time steps. We present ablation experiments and comparison with two existing baseline methods to highlight several design choices that we found to be essential to our model’s success.We test the proposed planning approach in a simulated highway driving environment, showing that by using the model, the agent can plan actions that mimic the interactive behavior of humans.