Dangerous driving behavior detection using video-extracted vehicle trajectory histograms

[国外交通期刊数据库] 作者:Zhijun Chen

摘要:Dangerous driving behavior detection can be used in video surveillance systemsto identify dangerous patterns, such as Abrupt Double Lane Change (ALC), Retrograde Driving (RD), and Illegal U-Turn (IT), for traffic design, traffic management, and law enforcement. The purpose of this study is to develop a detection method of dangerous driving behavior based on video surveillance. First, a novel method named trajectory histogram is proposed. A set of trajectory histograms (e.g., control points histogram and velocity histogram) is constructed to represent vehicle motion. Then, a frequently used feature selection method named Minimum Redundancy and Maximum Relevance (mRMR) is introduced to evaluate the most representative trajectory histograms for dangerous driving behavior detection. In addition, a hybrid algorithm, Particle SwarmOptimization-Support VectorMachine (PSO_SVM), is then developed to identify dangerous driving behavior. To validate the performance and effectiveness of the proposed method, several experiments are conducted. The results show that mRMR is better than other representative methods, namely Conditional Mutual Information Maximization (CMIM), Mutual Information Maximization (MIM), and ReliefF, for evaluating the most representative trajectory histograms. Based on the most representative trajectory histograms, the detection accuracy rate of dangerous driving behavior using PSO_SVM is superior to those of the most frequently used classifiers— Naïve Bayesian Classifier (NBC), k-Nearest Neighbor (kNN), and C4.5 decision tree. In addition, we find that the proposed method outperforms the two common approaches for dangerous driving behavior detection in video surveillance systems. Therefore, the proposed method can be widely applied to detect dangerous driving behavior in video surveillance systems.

Outsourced probe data effectiveness on signalized arterials

[国外交通期刊数据库] 作者:Elham Sharifi

摘要:This paper presents results of an I-95 Corridor Coalition sponsored project to assess the ability of outsourced vehicle probe data to provide accurate travel time on signalized roadways for the purposes of real-time operations as well as performance measures. The quality of outsourced probe data on freeways has led many departments of transportation to consider such data for arterial performance monitoring. From April 2013 through June of 2014, the University of Maryland Center for Advanced Transportation Technology gathered travel times from several arterial corridors within the mid-Atlantic region using Bluetooth traffic monitoring (BTM) equipment, and compared these travel timeswith the data reported to the I95 Vehicle Probe Project (VPP) froman outsourced probe data vendor. The analysis consisted of several methodologies: (1) a traditional analysis that used precision and bias speed metrics; (2) a slowdown analysis that quantified the percentage of significant traffic disruptions accurately captured in the VPP data; (3) a sampled distribution method that uses overlay methods to enhance and analyze recurring congestion patterns. (4) Last, the BTM and VPP data from each 24-hour period of data collection were reviewed by the research team to assess the extent to which VPP captured the nature of the traffic flow. Based on the analysis, probe data is recommended only on arterial roadways with signal densities (measured in signals per mile) up to one, and it should be tested and used with caution for signal densities between one and two, and is not recommended when signal density exceeds two.

Agents Toward Vehicle Routing Problem With Time Windows

[国外交通期刊数据库] 作者:PETR KALINA

摘要:We suggest an efficient polynomial algorithm for the vehicle routing problem with time windows based on agent negotiation. A generic agent decomposition of the problem is introduced featuring a clear separation between the local planning performed by the individual vehicles and the abstract global coordination achieved by negotiation—differentiating the presented algorithm from the classical centralized algorithms. The semantic of the abstract negotiation process is discussed, as well as the two alternative local planning strategies used by the individual vehicles. A relevant comparison to the state-of-the-art classical and agent-based algorithms is provided, missing from most previous agent-based studies. The algorithm performs slightly worse than the state-of-the-art classical algorithms but improves on all the previous comparable agent-based algorithms.

Learning Drivers' Behavior to Improve Adaptive Cruise Control

[国外交通期刊数据库] 作者:AVI ROSENFELD

摘要:Traditionally, vehicles have been considered as machines that are controlled by humans for the purpose of transportation. A more modern view is to envision drivers and passengers as actively interacting with a complex automated system. Such interactive activity leads us to consider intelligent and advanced ways of interaction leading to cars that can adapt to their drivers. In this article, we focus on the adaptive cruise control (ACC) technology that allows a vehicle to automatically adjust its speed to maintain a preset distance from the vehicle in front of it based on the driver's preferences. Although individual drivers have different driving styles and preferences, current systems do not distinguish among users. We introduce an approach to combine machine learning algorithms with demographic information and behavioral driver models into existing automated assistive systems. This approach can reduce the interactions between drivers and automated systems by adjusting parameters relevant to the operation of these systems based on their specific drivers and context of drive. We also learn when users tend to engage and disengage the automated system. This approach sheds light on the kinds of dynamics that users develop while interacting with automation and can teach us how to improve these systems for the benefit of their users. While generic packages such as Weka were successful in learning drivers' behavior exclusively based on the ACC's sensors, we found that improved learning models could be developed by adding information on drivers' demographics and a previously developed model about different driver types. We present the general methodology of our learning procedure and suggest applications of our approach to other domains as well.

Modeling the Impact of Traffic Information Acquisition From Mobile Devices During the Primary Tour of the Day

[国外交通期刊数据库] 作者:ATHENA TSIRIMPA

摘要:In this study the impact of real-time travel information acquired through mobile devices on traveler behavior is investigated. The study takes into consideration individuals' attitudes toward the use of smart mobile devices. The effect of traffic information acquisition in activity rescheduling is also considered. A case study was developed for the Athens Metropolitan Area, Greece, in 2009. A survey was conducted that included both revealed and stated preference data, as well as attitudes and perceptions of individual decision makers. This article focuses on the modeling and analysis of the stated preferences experiment conducted. An integrated error component latent variable model was estimated, predicting the probability of travelers' switching behavior as a function of travel information and traffic information seekers through mobile devices. The analysis of the results indicates that traffic information seekers through mobile devices are more likely to switch travel behavior, while traffic information appears to affect the rescheduling of individuals' activities that in many cases is also accompanied by mode or route switching.

A Stochastic Dilemma Zone Protection Algorithm Based on the Vehicles' Trajectories

[国外交通期刊数据库] 作者:PENGFEI LI

摘要:A common method of the dilemma zone (DZ) protection at intersections is to hold the green until the number of vehicles in DZ is lower than a threshold. Since the threshold is typically empirical and fixed, it cannot accommodate the dynamic and time-varying traffic patterns and therefore should be adjusted regularly. This article presents a new Markov-process-based DZ protection algorithm, which considers the number of vehicles in DZ (i.e., the state) over time to be a Markov process. At each time step, the algorithm first predicts the future states with the Markov state-transit matrix, then compares them with the current state to determine whether to end the green or not. In this way, the new end-green criterion is not the fixed threshold value but the current state and the prediction with the Markov state-transit matrix. Meanwhile, the Markov matrix is automatically updated whenever the new observed detected state transitions come in. The new algorithms were also evaluated in simulation and the simulation results showed that the new algorithm maintains reliable and effective protection in a dynamic traffic environment. At last, we find that the new algorithm performance can be further improved if the vehicle trajectories are precisely measured rather than estimated.

Intelligent Algorithms for Reducing Short-Term Traffic State Prediction Error in Active Traffic Management

[国外交通期刊数据库] 作者:JIE FANG

摘要:Macroscopic traffic flow models have been widely used to provide reliable prediction inputs for active transportation and demand management (ATDM). The representative models include METANET, FREFLOW, and KRONOS. In recent development of ATDM, METANET-based prediction models have started to more attention with their convenience of formulating several important macroscopic control variables, such as speed limit and ramp flow. However, applying METANET models on real-world observations (e.g., loop detector data) may experience significant performance issues, especially when applied to a corridor with heterogeneous links. Heterogeneous links refer to adjacent links that have different macroscopic traffic flow characteristics such as free flow speed and capacity. The heterogeneity is usually caused by changes in road geometry and environment, which commonly exist in urban freeway networks. This article provides an in-depth study to investigate the error mechanism of this problem and proposes several effective methods to improve the performance of METANET model over heterogeneous links. It is found in this article that applying some dynamic formulations to the convention term of the speed equation in METANET can significantly reduce the impact of link heterogeneity issue. The proposed method was evaluated using field loop detector data collected at I-894 corridor in Milwaukee, WI.

Using LIDAR to Validate the Performance of Vehicle Classification Stations

[国外交通期刊数据库] 作者:HO LEE

摘要:Vehicle classification is an important traffic parameter for transportation planning and infrastructure management. Vehicle classification stations depend on accurate calibration and validation to yield meaningful results. Operating agencies spend millions of dollars to deploy vehicle classification stations to collect classified count data, yet very few of these stations are ever subjected to a rigorous performance evaluation to ensure that they are reporting accurate data. To date such performance monitoring has been prohibitively labor intensive and prone to human error. To address this problem we develop a classification performance monitoring system to allow operating agencies to rapidly assess the health of their classification stations on a per vehicle basis. We eliminate most of the labor demands and instead deploy a portable nonintrusive vehicle classification system (PNVCS) to classify vehicles, concurrent with an existing classification station. This article uses a LIDAR-based PNVCS but our approach is compatible with many other PNVCSs. The processing requires several intermediate steps, developed herein, including synchronizing the independent clocks and matching observations of a given vehicle between the two classification systems. The performance monitoring methodology automatically compares the vehicle classification between the existing classification station and the PNVCS for each vehicle. If the two classification systems agree, the given vehicle is automatically taken as a success. A human only looks at the vehicle when the two systems disagree, and for this task we have developed tools to semi-automate the manual validation process, greatly increasing the efficiency and accuracy of the human user. Thus, the PNVCS must be accurate enough to be used as a baseline, as verified herein. The methodology is applied to over 21,000 vehicles from several permanent and temporary vehicle classification stations to evaluate the performance of axle and length-based classification stations. The automated process does the bulk of the work, with only 8％ of the vehicles requiring manual intervention. The user typically spent 3–5 seconds per vehicle reviewed, translating into only a few minutes to process the exceptions from all lanes over 1 hour of data. This approach offers a cost-effective solution to ensure that classification stations are providing accurate data, and for permanent classification stations the additional labor is a fraction of the cost to deploy the station in the first place.

Operational Analysis of a Freeway Variable Speed Limit System in St. Louis, Missouri

[国外交通期刊数据库] 作者:JALIL KIANFAR

摘要:Variable speed limit (VSL) systems have been more widely implemented in Europe compared to the United States. Studies have demonstrated positive safety impacts of such systems; however, there are very few studies that used actual deployment data to investigate the operational benefits of VSL systems. This article presents the operational impacts of a VSL deployment on Interstate 270 in the state of Missouri in the United States. Techniques such as parametric curve fitting, nonparametric methods, and other statistical tests were used to identify the changes between before and after traffic conditions. The effect of VSL on traffic performance was investigated at eight heavily congested locations. The two-dimensional Kolmogorov–Smirnov test results indicated that flow–occupancy diagram changes were statistically significant at seven out of eight locations. The slopes of flow–occupancy plots for over critical occupancies were found to be steeper after VSL. Slight changes in critical occupancy were observed after VSL implementation. However, the changes were inconsistent across locations, with some witnessing an increase and others witnessing a decrease. The maximum flow prior to breakdown (the prebreakdown flow) decreased at four locations and increased at four locations after VSL. The maximum flow after breakdown (the postbreakdown flow) decreased at three locations and increased at five locations after VSL. The average daily duration of congestion decreased at five locations and increased at three locations after VSL. Findings of this study help to develop VSL control algorithms that are more efficient in improving VSL traffic operations benefits.

Travel time estimation oriented freeway sensor placement problem considering sensor failure

[国外交通期刊数据库] 作者:Ning Zhu

摘要:Optimal sensor placement on freeway corridor is of great interest to transportation authorities. However, current traffic sensors are easily subject to various failures. Therefore, it is necessary to incorporate sensor failure into the optimal sensor placement model. In this article, a two-stage stochastic model is proposed for the purpose of travel time estimation on freeway corridor. To balance the effectiveness and reliability, a stochastic conditional value at risk (CVaR) model is also proposed. Since both models are too complicated, a customized genetic algorithm is developed. Numerical experiments show that considering sensor failure makes a significant performance improvement in the sensor placement pattern. Sensitivity analysis is also applied to investigate the impact of a number of allowable sensors and different traffic sensor failure probability.