Advanced Traveler Information Systems

This paper discusses two main themes. First, it investigates the formation of a spatiotemporal cognitive map (mental image) of a road network in travelers’ memory, which entails the travelers’ global conceptual understanding of congestion or the degree of crowding of the network. Second, it tries to investigate how latent learning of travelers from previous experiences shapes parts of the mental image, even for the parts of the network with which the travelers are unfamiliar. An experiment of route choice experiences was conducted among 90 participants in order to gain insight into the formation of a cognitive map and latent learning. In this experiment, the following independent variables are connected to the formation of a mental image of the network and the quality of the generalization of the unfamiliar parts of the network: (i) dispersion of the links’ travel time throughout the network, (ii) number of trips the traveler makes, (iii) traveler’s gender, (iv) traveler’s driving experiences, (v) traveler’s natural level of optimism or pessimism, (vi) salient or noticeable features on the network, and (vii) the presence of traffic signals. Several nonparametric (distribution-free) tests are employed to test the hypotheses. The results indicate that apart from the traveler’s gender and salient or noticeable features on the network, the considered factors significantly affect the degree of the recognizability of the network elements by travelers.

This paper is derived from one of our research mainstream, research theme: TRAFFIC & TRANSPORTATION PSYCHOLOGY ; TRAVEL PSYCHOLOGY

Abstract:
Travel time perception and learning play a central role in the modeling of day-to-day travel choice dynamics in traffic networks and have attracted the attention of many researchers, specifically for the analysis and operation of intelligent transportation systems and travel demand management scenarios. In this paper, a fuzzy learning model is proposed to capture the mechanism by which travelers update their travel time perceptions from one day to the next, taking into account their experienced travel times. In order to capture travelers’ mental representations of uncertain travel time involving imprecision and uncertainty, a combined artificial neural network and fuzzy logic (neuro-fuzzy) architecture called Adaptive-Network-based Fuzzy Inference System (ANFIS) is employed. This framework, which utilizes a set of fuzzy if-then rules, can serve as a basis for modeling the qualitative sides of travelers’ knowledge and reasoning processes. From the output of this study, the results of our laboratory-like experiment provide a good fit to the stated data of travelers’ behavior, and may reflect the fact that the neuro-fuzzy approach can be considered a promising method in learning and perception updating models. Finally, the proposed learning model is embedded in a microscopic event based simulation framework to evaluate its credibility within a day-to-day behavior of the traffic network. The results of the simulation, which converge to the equilibrium state of the test network, are finally presented, implying that the proposed perception-updating model operates properly.

A regional advanced traveler information system (ATIS)—the Regional Traveler Information Center (RTIC)—is being developed and implemented in western Massachusetts to provide real-time pre-trip and en-route travel time information to the public.  The main functions of the RTIC include traffic data collection, processing, and information dissemination.  The core of the data processing component is an arterial route travel time estimation and prediction model.  This model is being developed and tested based on the combination of traffic data from two types of traffic sensors—license plate readers (LPR) and microwave-ultrasonic-infrared (MUI) sensors—in the RTIC testbed, a 3.7-mile critical arterial corridor on State Route 9.  Individual route travel times captured by the LPR system and standard point traffic measures collected by the MUI sensors are integrated in order to overcome the inherent disadvantages of previous travel time estimation techniques based on either section-based or point-based data sources.  The model functions on two levels: (1) to provide current travel time information for on-trip route choice; and (2) to predict short-term travel time variations from the current travel times for pre-trip route planning.  Several separate functional modules are integrated into this model: (1) a heuristic LPR data filtering algorithm with underlying statistical components for validation of individual LPR travel time measurements; (2) a dynamic LPR data examination and aggregation module for providing aggregate LPR route travel time estimates; and (3) state space models built based on conditional independences between traffic variables, which are capable of correcting lagging LPR travel time estimates through monitoring the variation of MUI traffic flow data, and of deriving predictive route travel times in near-future time horizons in terms of the recent evolution of MUI traffic flow data.