Enhancing multi-scenario applicability of freeway variable speed limit control strategies using continual learning.

Variable speed limit (VSL) control benefits freeway operations through dynamic speed limit adjustment strategies for specific operation scenarios, such as traffic jams, secondary crash prevention, etc. To develop optimal strategies, deep reinforcement learning (DRL) has been employed to map the traffic operation status to speed limits with the corresponding control effects. Then, VSL control strategies were obtained based upon memories of these complex mapping relationships. However, under multi-scenario conditions, DRL trained VSL faces the challenge of performance decay, where the control strategy effects drop sharply for early trained "old scenarios". This so-called scenario forgetting problem is attributed to the fact that DRL would forget the learned old scenario mapping memories after new scenario trainings. To tackle this issue, a continual learning approach has been introduced in this study to enhance the multi-scenario applicability of VSL control strategies. Specifically, a gradient projection memory (GPM) based neural network parameter updating method was proposed to keep the mapping memories of old scenarios during new scenario trainings by imposing constraints on the direction of gradient updates for new tasks. The proposed method was evaluated using three typical freeway operation scenarios developed in the simulation platform SUMO. Experimental results showed that the continual learning approach has substantially reduced the performance decay in old scenarios by 17.76% (valued using backward transfer metrics). Furthermore, the multi-scenario VSL control strategies successfully reduced the speed standard deviation and average travel time by 28.77% and 7.25% respectively. Moreover, the generalization of the proposed continual learning based VSL approach were evaluated and discussed.

A reliability-based weather-responsive variable speed limit system to improve the safety of rural highways.

Weather-responsive Variable Speed Limit (WRVSL) systems treat speed limits as weather-dependent random variables, as opposed to the conventional static speed limits. This study (i) evaluates drivers' response to a fixed speed limit in different road-weather conditions, and (ii) proposes an effective approach to set WRVSLs, for rural divided highways located in extremely cold regions. Study data: road-weather, and speed data, collected from a rural highway (fixed speed limit = 110 km/h), are used to (i) estimate the 85th percentile speeds of population-level speed distributions, and (ii) develop WRVSLs based on the reliability theory. More specifically, the WRVSLs are set based on reliability: the probability of a speed being (i) likely complied by drivers, and (ii) adequate to avoid a rear-end collision. The study results reveal that merely 73 % of the drivers at the study site comply with the existing posted speed limit under normal road-weather conditions i.e., no precipitation and dry pavements. The reliability of the current speed limit is revealed to be approximately-one under normal road-weather conditions; thus, the current speed limit is perceived credible under such road-weather conditions. Yet, reliability of the current speed limit is substantially reduced in the presence of slight snow, and ice warning pavement conditions. A set of reliability-based WRVSLs ranging from 80 to 110 km/h is proposed. Jurisdictions experiencing extreme road-weather conditions may adapt the proposed methodology to effectively manage speed, particularly in rural highways under adverse road-weather conditions to enhance the probability of speed limits being safe and complied by drivers and as a result reduce crash propensity.

Active traffic management strategies for expressways based on crash risk prediction of moving vehicle groups.

Active traffic management (ATM) strategies are useful methods to reduce crash risk and improve safety on expressways. Although there are some studies on ATM strategies, few studies take the moving vehicle group as the object of analysis. Based on the crash risk prediction of moving vehicle groups in a connected vehicle (CV) environment, this study developed various ATM safety strategies, that is, variable speed limits (VSLs), ramp metering (RM), and coordinated VSL and RM (VSL-RM) strategies. VSLs were updated to minimize the crash risk of multiple moving vehicle groups in the next time interval, which is 1 min, and the updated speed limits were sent directly to the CVs in the moving vehicle group. The metering rate and RM opening time were determined using mainline occupancy, the crash risk of upcoming moving vehicle groups, and the predicted time at which moving vehicle groups arrived at the on-ramp. The VSL-RM strategy was used to simultaneously control and coordinate traffic flow on the mainline and ramps. These strategies were tested in a well-calibrated and validated micro-simulation network. The crash risk index and conflict count were utilized to evaluate the safety effects of these strategies. The results indicate that the ATM strategies improved the expressway safety benefits by 2.84-15.92%. The increase in CV penetration rate would promote the safety benefits of VSL and VSL-RM. Moreover, VSL-RM was superior to VSL and RM in reducing crash risk and conflict count.

Speed Management Strategy: Designing an IoT-Based Electric Vehicle Speed Control Monitoring System.

Road accidents represent the greatest public health burden in the world. Road traffic accidents have been on the rise in Rwanda for several years. Speed has been identified as a core factor in these road accidents. Therefore, understanding road accidents caused by excessive speeding is critical for road safety planning. In this paper, input and out pulse width modulation (PWM) was used to command the metal-oxide-semiconductor field-effect transistor (MOSFET) controller which supplied voltage to the motor. A structural speed control and Internet of Things (IoT)-based online monitoring system was developed to monitor vehicle data in a continuous manner. Two modeling techniques, multiple linear regression (MLR) and random forest (RF) models, were evaluated to find the best model to estimate the required voltage to be supplied to the motors in a particular zone. The built models were evaluated based upon the coefficient of determination R2. The RF performs better than the MLR as it reveals a higher R2 value and it is found to be 98.8%. Based on the results, the proposed method was proven to significantly reduce the supplied voltage to the motor and consequently increase safety.

A Bayesian modeling framework for crash severity effects of active traffic management systems.

Transportation agencies utilize Active traffic management (ATM) systems to dynamically manage recurrent and non-recurrent congestion based on real-time conditions. While these systems have been shown to have some safety benefits, their impact on injury severity outcomes is currently uncertain. This paper used full Bayesian mixed logit models to quantify the impact that ATM deployment had on crash severities. The estimation results revealed lower severities with ATM deployment. Marginal effects for ATM deployments that featured hard shoulder running (HSR) revealed lower likelihoods for severe and moderate injury crashes of 15.9 % and for minor injury crashes of 10.1 %. The likelihood of severe and moderate injury crashes and minor injury crashes reduced by 12.4 % and 8.33 % with ATM without HSR. The models were observed to be temporally transferable and had forecast error of 0.301 and 0.304 for the two models, revealing better performance with validation data. These results have implications for improving freeway crash risk at critical locations.

Trajectory-level fog detection based on in-vehicle video camera with TensorFlow deep learning utilizing SHRP2 naturalistic driving data.

Providing drivers with real-time weather information and driving assistance during adverse weather, including fog, is crucial for safe driving. The primary focus of this study was to develop an affordable in-vehicle fog detection method, which will provide accurate trajectory-level weather information in real-time. The study used the SHRP2 Naturalistic Driving Study (NDS) video data and utilized several promising Deep Learning techniques, including Deep Neural Network (DNN), Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and Convolutional Neural Network (CNN). Python programming on the TensorFlow Machine Learning library has been used for training the Deep Learning models. The analysis was done on a dataset consisted of three weather conditions, including clear, distant fog and near fog. During the training process, two optimizers, including Adam and Gradient Descent, have been used. While the overall prediction accuracy of the DNN, RNN, LSTM, and CNN using the Gradient Descent optimizer were found to be around 85 %, 77 %, 84 %, and 97 %, respectively; much improved overall prediction accuracy of 88 %, 91 %, 93 %, and 98 % for the DNN, RNN, LSTM, and CNN, respectively, were observed considering the Adam optimizer. The proposed fog detection method requires only a single video camera to detect weather conditions, and therefore, can be an inexpensive option to be fitted in maintenance vehicles to collect trajectory-level weather information in real-time for expanding as well as updating weather-based Variable Speed Limit (VSL) systems and Advanced Traveler Information Systems (ATIS).

Effects of connected vehicle-based variable speed limit under different foggy conditions based on simulated driving.

In response to developing and/or diminishing foggy conditions, the variable speed limit application in a connected vehicle environment (CV-VSL) can estimate and deliver recommended travel speeds to individual drivers, which can help to reduce crashes when visibility conditions change. This study aims to quantify the effectiveness of the CV-VSL application by exploring drivers' reactions to warnings (e.g., recommended travel speeds). In order to analyze the effectiveness of the CV-VSL application, a connected vehicle testing platform was established based on a driving simulator, and characteristics of the drivers' speed adjustments after receiving warnings were analyzed with respect to different levels of visibility (i.e., no fog, slight fog, and heavy fog). This study also examined the effect of warnings on drivers in different impact zones (i.e., clear zone, transition zone, and fog zone). Three indicators were identified: 1) speed at the end of the clear zone, 2) maximum deceleration rate in the transition zone, and 3) average speed reduction in the fog zone. Throughout the experiment, the relationship between speed adjustments and the level of visibility was explored. The results indicated that the CV-VSL application is effective in making drivers reduce travel speeds in all three types of zones. Furthermore, it appeared that the CV-VSL application could help manage travel speeds prior to vehicles entering the transition zone, and influence drivers' braking decisions upon encountering reduced visibility. It was also found that the CV-VSL application was more effective in heavy fog conditions than in light fog conditions. The connected vehicle testing platform based on the driving simulator provided a new method for evaluating the effectiveness of in-vehicle messaging generated by connected vehicle applications.

A novel method for imminent crash prediction and prevention.

A crash prediction and prevention method was proposed to detect imminent crash risk and help recommend traffic control strategies to prevent crashes. The method consists of two modules, the crash prediction module and the crash prevention module. The crash prediction module detects crash-prone conditions when the predicted crash probability exceeds a specified threshold. Then the crash prevention module would simulate the safety effect of traffic control alternatives and recommend the optimal one. The proposed method was demonstrated in a case study with variable speed limit (VSL). Results showed that the proposed crash prediction and prevention method could effectively detect crash-prone conditions and evaluate the safety and mobility impacts of various safety countermeasures.

Freeway crash risks evaluation by variable speed limit strategy using real-world traffic flow data.

The primary objective of this study is to evaluate the real-time crash risk of freeways by using real-world traffic flow data. The crash risk expressed as the potential crash likelihood is assessed under variable speed limit (VSL) and without VSL, in which both spatial correlation between different sites and temporal similarity are contained. Traffic flow data of Whitemud Drive network (WMD) in Canada is used to perform the relevant analysis, including VSL implementation analysis, traffic flow similarity analysis, crash risk and congestion analysis. Analytical results demonstrate that the average traffic flow under VSL schemes 1, 2, 3 and 4 are highly correlated from spatial-temporal perspective. The crash likelihoods and congestions under these VSL schemes are greatly improved. The best VSL control scheme, the most dangerous area and time, together with the most congested station of WMD are eventually determined. Subsequently, a t-test is employed to examine the significance of these results. t-Test results suggest that the improvement degree between crash risk and congestion under the best VSL control scheme show a difference, i.e., the best VSL control scheme can reduce the crash risk of moderate risk area more than high risk area, while it may have a larger melioration on the most congested area than the relatively uncongested area. Finally, these results are considered to have the potential reference in the mitigation of WMD traffic issues.

Traffic management: Assessing various countermeasures to improve detection failure of changes in speed limit signals.

Under certain circumstances, drivers fail to notice changes in electronic speed limits. A video-based study was performed to reveal which countermeasures would improve drivers' ability to detect changes in electronic speed limits. Countermeasures included leaving electronic signs blank prior to a speed limit change and adding motion signals by means of flashing amber lights or a wave. A video representing a motorway was shown repeatedly to 255 participants. They were instructed to press the space bar when detecting a change. The video was viewed 13 times before the speed limit changed. Results showed that leaving signs blank prior to the change instead of displaying speed limits continuously did not alter change detection, whereas flashers and waves eroded detection of the changed speed limit. This suggests that using flashers and waves to attract attention to electronic signs in fact decreases people's ability to process the information contained in the signs.

Reducing the risk of rear-end collisions with infrastructure-to-vehicle (I2V) integration of variable speed limit control and adaptive cruise control system.

OBJECTIVE: Adaptive cruise control (ACC) has been investigated recently to explore ways to increase traffic capacity, stabilize traffic flow, and improve traffic safety. However, researchers seldom have studied the integration of ACC and roadside control methods such as the variable speed limit (VSL) to improve safety. The primary objective of this study was to develop an infrastructure-to-vehicle (I2V) integrated system that incorporated both ACC and VSL to reduce rear-end collision risks on freeways. METHODS: The intelligent driver model was firstly modified to simulate ACC behavior and then the VSL strategy used in this article was introduced. Next, the I2V system was proposed to integrate the 2 advanced techniques, ACC and VSL. Four scenarios of no control, VSL only, ACC only, and the I2V system were tested in simulation experiments. Time exposed time to collision (TET) and time integrated time to collision (TIT), 2 surrogate safety measures derived from time to collision (TTC), were used to evaluate safety issues associated with rear-end collisions. The total travel times of each scenario were also compared. RESULTS: The simulation results indicated that both the VSL-only and ACC-only methods had a positive impact on reducing the TET and TIT values (reduced by 53.0 and 58.6% and 59.0 and 65.3%, respectively). The I2V system combined the advantages of both ACC and VSL to achieve the most safety benefits (reduced by 71.5 and 77.3%, respectively). Sensitivity analysis of the TTC threshold also showed that the I2V system obtained the largest safety benefits with all of the TTC threshold values. The impact of different market penetration rates of ACC vehicles in I2V system indicated that safety benefits increase with an increase in ACC proportions. CONCLUSIONS: Compared to VSL-only and ACC-only scenarios, this integrated I2V system is more effective in reducing rear-end collision risks. The findings of this study provide useful information for traffic agencies to implement novel techniques to improve safety on freeways.

Development of a variable speed limit strategy to reduce secondary collision risks during inclement weathers.

Inclement weather reduces traveler's sight distance and increases vehicle's stopping distance. Once a collision occurred during inclement weather and resulted in a slow traffic, approaching vehicles may not have adequate time to make emergency responses to the hazardous traffic, resulting in increased potentials of secondary collisions. The primary objective of this study is to develop a control strategy of variable speed limits (VSL) to reduce the risks of secondary collisions during inclement weathers. By analyzing the occurrence condition of secondary collision, the VSL strategy is proposed to dynamically adjust the speed limits according to the current traffic and weather conditions. A car-following model is modified to simulate the vehicle maneuvers with the VSL control. Two surrogate safety measures, based on the time-to-collision notion, are used to evaluate the control effects of VSL. Five weather scenarios are evaluated in simulation. The results show that the VSL strategy effectively reduces the risks of secondary collisions in various weather types. The time exposed time-to-collision (TET) is reduced by 41.45%-50.74%, and the time integrated time-to-collision (TIT) is reduced by 38.19%-41.19%. The safety effects are compared to those with a previous VSL strategy. The results show that in most cases our strategy outperforms the previous one. We also evaluate how driver's compliance to speed limit affects the effectiveness of VSL control.