James A Misener

PATH Investigations in Vehicle-Roadside Cooperation and Safety: A Foundation for Safety and Vehicle-Infrastructure Integration Research

The California Partners for Advanced Transit and Highways (PATH) has been active in researching vehicle-roadside cooperation since 1988. Through the years PATHs work in automation has spawned a considerable body of research in related topics of safety and vehicle-infrastructure cooperation. These topics are summarized by exemplar projects in areas of heavy vehicle driver assistance, cooperative forward collision avoidance, intersection safety and vehicle-infrastructure integration. While this review is by no means comprehensive, it illustrates the past, present and future of PATH and underscores our contribution toward and vision for future research in these topics

On-board road condition monitoring system using slip-based tyre-road friction estimation and wheel speed signal analysis

Abstract This paper presents an on-board road condition monitoring system. The road condition is continuously evaluated in terms of slipperiness and coarseness and is classified into four grades, normal (μmax ≤ 0.5), slippery (0.3 ≤ μmax< 0.5), very slippery (μmax< 0.3), and rough surface (gravel). A non-linear curve fitting technique is adopted to estimate the maximum tyre-road friction coefficient using the so-called ‘magic formula’. The characteristic of the relationship between friction coefficient and slip, i.e. the value of maximum friction coefficient μmax varies significantly with different surfaces, but its corresponding slip value λmax does not vary much, is exploited in the road condition classification algorithm. For surface coarseness analysis, a separate classifier based on the variance of filtered wheel speed signal is implemented. Experimental results demonstrate the feasibility of the road condition monitoring system for detecting slippery and rough road surfaces in close to real-time. In addition, the proposed slip-based friction estimation algorithm has the merits of robustness to vehicle-tyre variance and easy calibration as opposed to past slip-based friction estimation approaches in the literature.

Prediction of Red Light Running Based on Statistics of Discrete Point Sensors

A probabilistic model is proposed to predict red light running (RLR) for collision avoidance systems at arterial signalized intersections. This RLR predictor consists of an arrival time estimator and a statistical predictor of vehicle stop-and-go maneuvers with two discrete point sensors (capable of measuring speed). In addition, unlike most prediction models, which are designed to minimize mean errors, this model identifies two types of error: the false alarm and the missed report. The capability of distinguishing these two types of error is crucial to the effectiveness of RLR-related collision avoidance systems. Therefore, the Neyman–Pearson criterion is employed: it keeps the false-alarm rate lower than a given threshold while at the same time minimizing the probability of missing error. To quantify the trade-off between these two types of error in the system design, a system operating characteristics (SOC) function is defined. The system parameters are determined by using an offline supervised parameter-setting procedure in which training data are collected from a field intersection in the San Francisco Bay Area in California with Autoscope video cameras. Effectiveness of the proposed model and its prediction algorithm are demonstrated by the collected field data. The theoretical system performance predicted by the SOC curve is matched with the evaluated performance by means of data collected from field intersections. For example, at a preset false-alarm rate of 3%, the correct prediction rate of RLR for three approaches of the field intersection ranges from 63% to 80%.

BENEFIT EVALUATION OF CRASH AVOIDANCE SYSTEMS

A five-layer hierarchy to integrate models, data, and tools is proposed for benefits assessment and requirements development for crash avoidance systems. The framework is known as HARTCAS: Hierarchical Assessment and Requirements Tools for Crash Avoidance Systems. The analysis problem is multifaceted and large-scale. The driving environment is diverse and uncertain, driver behavior and performance are not uniform, and the range of applicable collision avoidance technologies is wide. Considerable real-world data are becoming available on certain aspects of this environment, although the collection of experimental data on other aspects is constrained by technological and institutional issues. Therefore, analyses of collision avoidance systems are to be conducted by collecting data on nominal operating conditions to the greatest extent possible and by using such data to build models for analysis of the rare, abnormal conditions. HARTCAS provides a framework within which to structure the collection and use of such knowledge. It is described in general terms, and its use is illustrated by analysis of a forward collision warning system. How to quantify the relationships between the effectiveness of a warning and the probability that the warning is a nuisance is shown. System benefits are also quantified.

Digital map as a virtual sensor – dynamic road curve reconstruction for a curve speed assistant

While digital map has been applied to many advanced driver assistance system applications, one critical attribute – road/lane curvature—is not available in the existing digital maps. Curvature derivation using splines is a well-known method in the computer graphics modelling community; however, it was also found that the direct application of the spline method is often not appropriate for real-time safety applications due to the resultant discontinuities in the curvature estimates and the lack of robustness against map data errors [CAMP, Enhanced digital mapping project final report, United States Department of Transportation, Washington, DC, 2004]. This paper presents a method to reconstruct road curvature attributes in real time using digital map data based on the proposed circle centre search and circle selection algorithms. Simulation and experimental results demonstrated that the proposed method can deliver curvature estimates that meet the desired accuracy. The method was also applied to the curve over-speed warning system in an on-going research project: Vehicle Infrastructure Integration—California. Some preliminary experimental results are also presented in this paper.

Observations of Driver Time Gap Acceptance at Intersections in Left-Turn Across-Path-Opposite-Direction Scenarios

Intersection collision warning systems can potentially reduce the number of collisions and associated losses. A critical design aspect of these systems is the selection of warning criteria, which represent a set of conditions and parameters under which the decision and the timing to issue warnings are determined. Proper warning criteria allow the generation of timely signals for drivers while minimizing false and nuisance alarms. The paper describes the development of a methodology to observe and analyze the selection of time gaps exhibited by driver behaviors in a real-world setting. The data collection procedures and analysis techniques are explained for left-turn across-path-opposite-direction scenarios, which constitute more than a quarter of crossing path crashes at intersections. Exemplar data sets from an urban, signalized intersection are used to illustrate methods of deriving time gap acceptance behaviors. The extracted information can serve as the basis for selecting gap acceptance thresholds in warning criteria, and the demonstrated methodology can be applied in the development of intersection collision warning systems.

Emergence of a Cognitive Car-Following Driver Model: Application to Rear-End Crashes with a Stopped Lead Vehicle

Rear-end crashes are a major roadway safety problem, and the potential of crash countermeasures to address this has long been recognized. High-frequency or severe-consequence scenarios are focused on the general lead-vehicle-not-moving (LVNM) case and specific crash scenarios. Operating scenarios are identified, and frequencies are assessed. From these, a small number of prevalent LVNM crash scenarios are identified as the focus for subsequent model development and crash counter-measure efforts. These scenarios suggest nominal atmospheric, roadway, lighting, vehicle, and driver conditions in designing cost-effective safety features to avoid LVNM rear-end crashes. From this, emergent models for cognitive car following are developed, based on fusing current knowledge. This will serve as a foundation for further model development efforts as well as for future human-factors experiments.

Active Highways (Position Paper)

Highways are an essential component of our society because they are critical to quality of life and to local and national economies. Under good conditions, highways provide a safe and efficient route for people and goods to reach their destinations. However, as a direct consequence of their use, traffic congestion is ever-increasing, undermining the ability of highways to adequately provide an acceptable quality of service. It has become imperative for highway traffic to provide the same time guarantee quality as other transportation methods such as air and rail travel, while maintaining the convenience of flexible scheduling and destination for the individual traveler. In this position paper, we propose Active Highways, a fundamental departure from todaypsilas highway traffic management approaches that shifts the highway paradigm from a transportation infrastructure that monitors and controls traffic at the aggregate level, to a computer-based service that operates at the level of individual vehicles. In this sense, highways will become active managers of their own traffic similar to air traffic control. In our vision, future highways and future vehicles will communicate with one another, making the highway system aware of the driverspsila travel plans and allowing it to cooperate with and actively instruct the driver on achieving them. In particular, Active Highways will allow drivers to reserve slots in special high-priority intelligent lanes. This fine-grained traffic management model will guarantee travel time bounds, handle exceptions and enforce global community and environmental policies using real-time information from vehicle- and infrastructure-based sensors.

Sensor-friendly vehicle and roadway systems

Sensor-friendly vehicle and roadway systems consist of complementary signal sensor and reflector or transmitter technologies, which provide information about the threat of a collision. These technologies can be composed into cooperative collision avoidance systems, which can supplement or replace single vehicle-based systems. Experiments were run on the four most promising technologies to determine their performance and reliability; the four technologies were passive license plates with enhanced radar return, roadside obstacle-mounted radar-reflecting corner cubes, fluorescent paint for lane and obstacle marking, and light emitting diode brake-light messaging. These technologies all focus on improving the signal-to-noise ratio of the collision avoidance sensor. We believe that experimental results indicate that further proof-of-concept refinements are needed, but in general these systems represent technologically sound, cooperative vehicle-roadway components and that sensor friendly systems could eventually translate into a significant benefit in terms of lives saved.

Threat assessment of traffic moving toward a controlled intersection

This paper presents an approach for threat assessment of traffic flows in a controlled intersection, based on empirical traffic data observed in an urban environment. The real-world traffic data collected from an observation study allows a glimpse of representative traffic patterns. The analysis of such data yields insights into the behaviors of traffic streams as a function of traffic signal phases. The determination of these critical parameters provides a practical basis for the developments of safety algorithms for threat assessment, and it serves as a grounded foundation for assessing sensing requirements for an effective intersection decision-support system.