Spatial-Temporal Flows-Adaptive Street Layout Control Using Reinforcement Learning

Complete streets scheme makes seminal contributions to securing the basic public right-of-way (ROW), improving road safety, and maintaining high traffic efficiency for all modes of commute. However, such a popular street design paradigm also faces endogenous pressures like the appeal to a more balanced ROW for non-vehicular users. In addition, the deployment of Autonomous Vehicle (AV) mobility is likely to challenge the conventional use of the street space as well as this scheme. Previous studies have invented automated control techniques for specific road management issues, such as traffic light control and lane management. Whereas models and algorithms that dynamically calibrate the ROW of road space corresponding to travel demands and place-making requirements still represent a research gap. This study proposes a novel optimal control method that decides the ROW of road space assigned to driveways and sidewalks in real-time. To solve this optimal control task, a reinforcement learning method is introduced that employs a microscopic traffic simulator, namely SUMO, as its environment. The model was trained for 150 episodes using a four-legged intersection and joint AVs-pedestrian travel demands of a day. Results evidenced the effectiveness of the model in both symmetric and asymmetric road settings. After being trained by 150 episodes, our proposed model significantly increased its comprehensive reward of both pedestrians and vehicular traffic efficiency and sidewalk ratio by 10.39%. Decisions on the balanced ROW are optimised as 90.16% of the edges decrease the driveways supply and raise sidewalk shares by approximately 9%. Moreover, during 18.22% of the tested time slots, a lane-width equivalent space is shifted from driveways to sidewalks, minimising the travel costs for both an AV fleet and pedestrians. Our study primarily contributes to the modelling architecture and algorithms concerning centralised and real-time ROW management. Prospective applications out of this method are likely to facilitate AV mobility-oriented road management and pedestrian-friendly street space design in the near future.

Advances in pedestrian travel monitoring: Temporal patterns and spatial characteristics using pedestrian push-button data from Utah traffic signals

In this study, we advanced pedestrian travel monitoring using a novel data source: pedestrian push-button presses obtained from archived traffic signal controller logs at more than 1,500 signalized intersections in Utah over one year. The purposes of this study were to: (1) quantify pedestrian activity patterns; (2) create factor groups and expansion/adjustment factors from these temporal patterns; and (3) explore relationships between patterns and spatial characteristics. Using empirical clustering, we classified signals into five groups, based on normalized hourly/weekly counts (each hour’s proportion of weekly totals, or the inverse of the expansion factors), and three clusters with similar monthly adjustment factors. We also used multinomial logit models to identify spatial characteristics (land use, built environment, socio-economic characteristics, and climatic regions) associated with different temporal patterns. For example, we found that signals near schools were much more likely to have bimodal daily peak hours and lower pedestrian activity during out-of-school months. Despite these good results, our hourly/weekday patterns differed less than in past research, highlighting the limits of existing infrastructure for capturing all kinds of activity patterns. Nevertheless, we demonstrated that signals with push-button data are a useful supplement to existing permanent counters within a broader pedestrian traffic monitoring program.

Seamless Wayfinding by a Deafblind Adult on an Urban College Campus: A Case Study on Wayfinding Performance, Information Preferences, and Technology Requirements

This article reports on an empirical evaluation of the experience, performance, and perception of a deafblind adult participant in an experimental case study on pedestrian travel in an urban environment. The case study assessed the degree of seamlessness of the wayfinding experience pertaining to routes that traverse both indoor and outdoor spaces under different modalities of technology-aided pedestrian travel. Specifically, an adult deafblind pedestrian traveler completed three indoor/outdoor routes on an urban college campus using three supplemental wayfinding support tools: a mobile application, written directions, and a tactile map. A convergent parallel mixed-methods approach was used to synthesize insights from a pre-travel questionnaire, route travel video recordings, post-travel questionnaire, and post-travel interview. Our results indicate that wayfinding performance and confidence differed considerably between the three wayfinding support tools. The tactile map afforded the most successful wayfinding and highest confidence. Wayfinding performance and confidence were lowest for the mobile application modality. The simplicity of use of a wayfinding tool is paramount for reducing cognitive load during wayfinding. In addition, information that does not match individual, user-specific information preferences and needs inhibits wayfinding performance. Current practice pertaining to the representation of digital spatial data only marginally accounts for the complexity of pedestrian human wayfinding across the gamut of visual impairment, blindness, and deafblindness. Robust orientation and mobility training and skills remain key for negotiating unexpected or adverse wayfinding situations and scenarios, irrespective of the use of a wayfinding tool. A substantial engagement of the deafblind community in both research and development is critical for achieving universal and equitable usability of mobile wayfinding technology.

Squeaky wheels: Missing data, disability, and power in the smart city

Data about the accessibility of United States municipalities is infrastructure in the smart city. What is counted and how, reflects the sociotechnical imaginary (norms and values) of a time or place. In this paper we focus on features identified by people with disabilities as promoting or hindering safe pedestrian travel. We use a regionally stratified sample of 178 cities across the United States. The municipalities were scored on two factors: their open data practices (or lack thereof), and the degree to which they cataloged the environmental features that persons with disabilities deemed critical for safe movement through urban spaces. In contradiction to the dominating narrative of too much data and not enough analyses, we find that when it comes to data points that might be useful to persons with disabilities, data are lacking. This data gap has consequences both politically and materially—on one hand data could help enforce compliance with the Americans with Disabilities Act, on the other they would allow for safe route planning. We find that reading these data formats and collection patterns from the perspective of critical disability studies—particularly those whose work disrupts notions of “normal” —helps answer questions about potential benefits and harms of data practices. This lens has the potential to promote analysis that is as disruptive to injustices as it is practical.

Travel and us: The impact of mode share on sentiment using geo-social media and GIS

Commute stress is a serious health problem that impacts nearly everyone. Considering that microblogged geo-locational information offers new insight into human attitudes, the present research examined the utility of geo-social media data for understanding how different active and inactive travel modes affect feelings of pleasure, or displeasure, in two major U.S. cities: Chicago, Illinois and Washington D.C. A popular approach was used to derive a sentiment index (pleasure or valence) for each travel Tweet. Methodologically, exploratory spatial data analysis (ESDA) and global and spatial regression models were used to examine the geography of all travel modes and factors affecting their valence. After adjusting for spatial error associated with socioeconomic, environmental, weather, and temporal factors, spatial autoregression models proved superior to the base global model. The results showed that water and pedestrian travel were universally associated with positive valences. Bicycling also favorably influenced valence, albeit only in D.C. A noteworthy finding was the negative influence temperature and humidity had on valence. The outcomes from this research should be considered when additional evidence is needed to elevate commuter sentiment values in practice and policy, especially in regards to active transportation.

Travel and us: The impact of mode share on sentiment using geo-social media and GIS

Commute stress is a serious health problem that impacts nearly everyone. Considering that microblogged geo-locational information offers new insight into human attitudes, the present research examined the utility of geo-social media data for understanding how different active and inactive travel modes affect feelings of pleasure, or displeasure, in two major U.S. cities: Chicago, Illinois and Washington D.C. A popular approach was used to derive a sentiment index (pleasure or valence) for each travel Tweet. Methodologically, exploratory spatial data analysis (ESDA) and global and spatial regression models were used to examine the geography of all travel modes and factors affecting their valence. After adjusting for spatial error associated with socioeconomic, environmental, weather, and temporal factors, spatial autoregression models proved superior to the base global model. The results showed that water and pedestrian travel were universally associated with positive valences. Bicycling also favorably influenced valence, albeit only in D.C. A noteworthy finding was the negative influence temperature and humidity had on valence. The outcomes from this research should be considered when additional evidence is needed to elevate commuter sentiment values in practice and policy, especially in regards to active transportation.

Pedestrian Traffic Signal Data Accurately Estimates Pedestrian Crossing Volumes

Existing methods of pedestrian travel monitoring are generally inefficient for collecting pedestrian data in many locations over long time periods. In this study, we demonstrate the validity of using a novel and relatively ubiquitous big data source—pedestrian data from high-resolution traffic signal controller logs—as a way of estimating pedestrian crossing volumes. Every time a person presses a pedestrian push button or a pedestrian call is registered at a signal, this information can be logged and archived. To validate these pedestrian signal data against observed pedestrian counts, we recorded over 10,000 h of video at 90 signalized intersections in Utah, and counted around 175,000 people walking. For each hour and crossing, we compared these observed counts to measures of pedestrian activity calculated from traffic signal data, using a set of five simple piecewise linear and quadratic regression models. Overall, our results show that traffic signal data can be successfully used to estimate pedestrian crossing volumes with good accuracy: model-predicted volumes were strongly correlated (0.84) with observed volumes and had a low mean absolute error (3.0). We also demonstrate how our models can be used to estimate annual average daily pedestrian volumes at signalized intersections and identify high pedestrian volume locations. Transportation agencies can use pedestrian signal data to help improve pedestrian travel monitoring, multimodal transportation planning, traffic safety analyses, and health impact assessments.