mehdi ghatee

In this paper, we propose a single camera-based driver assistance system for blind spots that is selfsupervised and learn without labeled data. The proposed system is developed based on deep neural network and uses gray scale image and optical flow as input. The strength of this paper compared to similar articles is the processing of image sequence information to assess the risk of accident due to moving objects in blind spots. In this paper, based on detection and tracking of the required objects, the existence of vehicles and its relative speed are estimated and the accident can be predicted for the blind spots of the car. The fusion of image with optical flow and features extracted using a deep neural network has increased robustness of the proposed system. In the proposed application of the article, the camera is mounted on the side mirror of the vehicle and is estimated to have a 60% accuracy, the risk of accidentally turning on the sides or changing lanes.

Influence maximization (IM) is a challenging problem in social networks to identify initial spreaders with the best influence on other nodes. It is a need to solve this problem with the minimum diffusion time and the most coverage on the communities. However, the spreaders are rarely dependent on diffusion models. A recent research [N. Binesh, M. Ghatee, Distance-Aware Optimization Model for Influential Nodes Identification in Social Networks with Independent Cascade Diffusion, Information Sciences, 581 (2021) 88-105] proposed DASF algorithm for spreaders selection by the Independent Cascade (IC) diffusion model. Here, we present a new optimization model to find spreaders under Linear Threshold (LT) diffusion model. LT is one of the most important models to imitate the behavior of influence propagation in social networks. Our model is a quadratic programming problem based on Laplacian-Plus matrix. We derive its solution by some principal eigenvectors of Laplacian-Plus matrix. We organize the solution process as DALT algorithm. Without community detection, it can identify the spreaders with maximum inter-communities distance, minimum intra-communities distance, and the most significant degrees. By considering various well-known social networks, we show that DALT provides brilliant results and overcomes other local and global spreader finders, especially in social networks with community structures.

This paper proposes an integrated system to control ramps and adjust variable speed limits. It includes three essential modules to predict the starting time of congestion and a fuzzy controller to determine the parameters and a model predictive control. An Apriori algorithm that is a powerful tool for frequent pattern mining is used in the first module. The proposed system is neither sensitive to the traffic distribution nor computationally intensive. Two traffic simulators of Aimsun and CTMSIM are applied to validate the results. Compared with the most recent algorithms, including Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM), this system improves prediction accuracy up to 2.63%. The results of ramp metering and variable speed limit subsystems are also promising. The embedded controller shows 0.6% and 4% overall and rush hour improvement in the total travel time.

Nowadays, we face a huge number of high-dimensional data in different applications and technologies. To tackle the challenge, various feature selection methods have been recently proposed for reducing the computational complexity of the learning algorithms as well as simplifying the learning models. Maintaining the geometric structures and considering the discriminative information in data are two important factors that should be borne in mind particularly for unsupervised feature selection methods. In this paper, our aim is to propose a new unsupervised feature selection approach by considering global and local similarities and discriminative information. Furthermore, this unsupervised framework incorporates cluster analysis to consider the underlying structure of the samples. Moreover, the correlation of features and clusters is computed by an -norm regularized regression to eliminate the redundant and irrelevant features. Finally, a unified objective function is presented as well as an efficient iterative optimization algorithm to solve the corresponding problem with some theoretical analysis of the convergence and the complexity of the algorithm. We compare the proposed approach with the state-of-the-art method based on clustering results on the various standard datasets including biology, image, voice, and artificial data. The experimental results have presented the strength and performance improvement of the proposed method by outperforming the well-known methods.

This paper proposes a smartphone-based warning system to evaluate the risk of a motor vehicle for vulnerable pedestrians (VP). The acoustic sensors are embedded in the roadside to receive vehicle sounds and they are classified into heavy vehicles, light vehicles with low speed, light vehicles with high speed, and no vehicle classes. For this aim, we extract new features by Mel-frequency Cepstrum Coefficients (MFCC) and Linear Predictive Coefficients (LPC) algorithms. We use different classification algorithms and show that MLP neural network achieves at least 96.77% accuracy criterion. To install this system, directional microphones are embedded on the roadside and the risk is classified. Then, for every microphone, a danger area is defined and the warning alarms have been sent to every VPs’ smartphones covered in this danger area.

Under dynamic conditions on bridges, we need a real-time management. To this end, this paper presents a rule-based decision support system in which the necessary rules are extracted from simulation results made by Aimsun traffic micro-simulation software. Then, these rules are generalized by the aid of fuzzy rule generation algorithms. Then, they are trained by a set of supervised and the unsupervised learning algorithms to get an ability to make decision in real cases. As a pilot case study, Nasr Bridge in Tehran is simulated in Aimsun and WEKA data mining software is used to execute the learning algorithms. Based on this experiment, the accuracy of the supervised algorithms to generalize the rules is greater than 80%. In addition, CART decision tree and sequential minimal optimization (SMO) provides 100% accuracy for normal data and these algorithms are so reliable for crisis management on bridge. This means that, it is possible to use such machine learning methods to manage bridges in the real-time conditions.

Nowadays collison warning systems are based on sharing position data between vehicles. The most common solution is using Dedicated Short Range Communication(DSRC). But this technology has met a lot of limitations. The spread of cellular network technology and it’s wide area coverage can make cellular network a proper alternative for DSRC. In this paper,first the collision senarios have been reviewed and their use cases are specified. It has been showed that 100 msec time delay is not essential in all collision senarios. Then 3G and 4G/LTE are evaluated for using as the main network in connected vehicles. The results show that 4G/LTE has the proper potential for using in vehicular network instead of DSRC. The time delay in 4G/LTE was below 100 msec in 98.14 per cent of samples. Also 3G had the mean time delay of 250 msec and can only be used in senarios that the vehicle’s speed are low and the distance between vehicles is high.

Nearly one third of accident casualties is related to pedestrians. Therefore pedestrian safety is one of the important transportation issues. For this purpose collision warning systems are mainly used. In this research, the necessary warnings are provided by computing and sending the vehicle and pedestrian smartphone positions, vehicle and pedestrian speeds, vehicle and pedestrian directions and vehicle acceleration to new fuzzy system. The proposed system includes three phases. In activation phase, the system according to position and trajectory of the vehicle and the pedestrians around the vehicle, is activated for the pedestrians in the activation area. In the second phase, detected pedestrians are monitored momentary to extract the effective collision parameters. And in the third phase, red level warning or yellow level warning is provided by fuzzification of effective parameters and accident risk analysis. The results show that activation of the system has 86 percent accuracy and warning or no warning situations are correctly detected in 95 percent of the cases. Also yellow warning and red warning accuracy was 62 and 100 percent.

Optimal usage of transportation equipments are very important issue in many countries which use ITS. Tunnel is one of the transportation structures which has many transportation systems as Jet fans، LED، CCTV،… so there is a need to have a control room in tunnel، which helps using these sensors، and make decisions in different critical situations rapidly. In this paper operation of control room to manage tunnel intelligent transportation system، is mentioned and then based on knowledge engineering rule extractions are discussed. These rules are used as a database for expert system. Then they used for tunnel management. In knowledge engineering، learnable algorithms are run، and near extracting of rules، they made a report for tunnel administrator. As a result Naive bayes decision tree with lower process time and higher accuracy made a best result for Niayesh tunnel in Tehran.

In this paper, a multi-perspective decision support system (MP-DSS) to design hierarchical public transportation network is developed. Since this problem depends on different perspectives, MP-DSS consists of two sub-systems with macro and micro sub-systems based on travel information, land use and expert knowledge. In the micro sub-system, two sub-modules are developed considering origin-destination demand matrix and attractive places to travel. In the first sub-system, based on traffic assignment models, the bus corridors can be extended and by the second approach, connectivity between attractive places can be provided by new bus lanes. Multi-commodity flow problem and spanning tree problem are used in these two sub-modules to assign the public services to the corresponding networks. The corridors obtained from these sub-modules are evaluated by experts board module. These corridors are used to extend bus rapid transit (BRT), exclusive bus lanes between multiple districts and shuttle buses for trips inside of district. A prototype of MP-DSS is developed to illustrate the results on Tehran network. The most important contribution of this paper is to generalize the different mathematical models with land use and expert knowledge which substantially improves the results of network designing problem.

This paper deals with linear programming problem with interval numbers as coefficients to exhibit with uncertainty. Since, the set of common intervals is not a field, we define generalized interval numbers to produce an algebraic interval field and on this field, we propose principle of uncertainty traverse instead of extension principle which permits to define operators on intervals exactly similar to the same operators on real numbers. In addition, we apply a total order on this field to transform interval linear programming into a traditional problem. The proposed order can be extended either pessimistically or optimistically. The numerical experiments are given to demonstrate the efficiency of the proposed scheme in comparison with the previous established works. The approach in this paper can be generalized to fuzzy linear programming problems taking the fuzzy cuts into account.

This paper proposes a new hybrid method namely SA-IP including simulated annealing and interior point algorithms to find the optimal toll prices based on level of service (LOS) in order to maximize the mobility in urban network. By considering six fuzzy LOS for flows, the tolls of congested links can be derived by a bi-level fuzzy programming problem. The objective function of the upper level problem is to minimize the difference between current LOS and desired LOS of links. In this level, to find optimal toll, a simulated annealing algorithm is used. The lower level problem is a fuzzy flow estimator model with fuzzy link costs. Applying a famous defuzzification function, a real-valued multi-commodity flow problem can be obtained. Then a polynomial time interior point algorithm is proposed to find the optimal solution regarding to the estimated flows. In pricing process, by imposing cost on some links with LOS F or E, users incline to use other links with better LOS and less cost. During the iteration of SA algorithm, the LOS of a lot of links gradually closes to their desired values and so the algorithm decreases the number of links with LOS worse than desirable LOS. Sioux Falls network is considered to illustrate the performance of SA-IP method on congestion pricing based on different LOS. In this pilot, after toll pricing, the number of links with LOS D, E and F are reduced and LOS of a great number of links becomes C. Also the value of objective function improves 65.97% after toll pricing process. It is shown optimal toll for considerable network is 5 dollar and by imposing higher toll, objective function will be worse.

In this paper, a novel multi-objective bus stop location model is proposed, which considers not only the coverage of demand and minimization of access time but also the necessities of suitable stops for transit network design phase. Three objective functions are considered including minimizing (I) sum of the total access distance (time), (II) the weighted combination of stops, and (III) the number of stops. A sum-weighted method is used to solve the proposed multi-objective model considering the different scenarios of weights. A detailed analysis is carried out Tehran CBD to generate sensible stops results.

A route guidance system is one of the most important services of advanced traveler information systems (ATIS) for drivers,which aims to inform the state of the highway’s traffic to a group of drivers using variable message sign (VMS), web sites, medias and etc. Dynamic travel time information is one of the major components of the dynamic route guidance system which can increase the highway networks reliability, lighten the traffic flow and create dynamic route guidance for drivers. In this paper different methods are reviewed and the results of neuro-fuzzy neural networks and perceptron neural networks are compared. Comparing the results of these methods for predicting the data collected from the Chamran to Niyayesh ramp in Tehran highway network, shows that fuzzy-neural approach can achieve more accurate results.

In recent years, due to increase in thed volume of traffic in large cities, estimating supply and demand estimates between production areas and to attract muchorigins and destinations attention has been under focusbeen. Since finding such the values using census between from different regions of the census metropolitan area network is very costly and time consuming, tThe use of traffic counters to determine the volume of traffic on network links is highly importantregarded. But in order to determine the minimum number and location of the counters locate different models is have been presented. One of the tools used in this context, is the vehicle license plate recognition cameras. In this paper, based on the performance of the cameras, the new two models is have been proposed to locate cameras. In tThe first model tries maximization of path coverage by adding a budget constraint to the location problem, to maximize coverage of paths is given. TIn the second model tries, simultaneous the maximization of the number of the scan lines and minimization ofing the number of the cameras to find the simultaneously answer the question to be determined. The simulation results study of both the problems on west Tehran highway network has been presented using both the AIMMS software package on the western part of Tehran's highway network is provided and performance of each model is has been checkedevaluated.