arXiv subjects

Nir Fulman

Publications and source records attributed to Nir Fulman.

7 recordsLinked to original sources

arXiv2024

Deep Learning Enhanced Road Traffic Analysis: Scalable Vehicle Detection and Velocity Estimation Using PlanetScope Imagery

This paper presents a method for detecting and estimating vehicle speeds using PlanetScope SuperDove satellite imagery, offering a scalable solution for global vehicle traffic monitoring. Conventional methods such as stationary sensors and mobile systems like UAVs are limited in coverage and constrained by high costs and legal restrictions. Satellite-based approaches provide broad spatial coverage but face challenges, including high costs, low frame rates, and difficulty detecting small vehicles in high-resolution imagery. We propose a Keypoint R-CNN model to track vehicle trajectories across RGB bands, leveraging band timing differences to estimate speed. Validation is performed using drone footage and GPS data covering highways in Germany and Poland. Our model achieved a Mean Average Precision of 0.53 and velocity estimation errors of approximately 3.4 m/s compared to GPS data. Results from drone comparison reveal underestimations, with average speeds of 112.85 km/h for satellite data versus 131.83 km/h from drone footage. While challenges remain with high-speed accuracy, this approach demonstrates the potential for scalable, daily traffic monitoring across vast areas, providing valuable insights into global traffic dynamics.

cs.CV↗

arXiv2024

Distortions in Judged Spatial Relations in Large Language Models

We present a benchmark for assessing the capability of Large Language Models (LLMs) to discern intercardinal directions between geographic locations and apply it to three prominent LLMs: GPT-3.5, GPT-4, and Llama-2. This benchmark specifically evaluates whether LLMs exhibit a hierarchical spatial bias similar to humans, where judgments about individual locations' spatial relationships are influenced by the perceived relationships of the larger groups that contain them. To investigate this, we formulated 14 questions focusing on well-known American cities. Seven questions were designed to challenge the LLMs with scenarios potentially influenced by the orientation of larger geographical units, such as states or countries, while the remaining seven targeted locations were less susceptible to such hierarchical categorization. Among the tested models, GPT-4 exhibited superior performance with 55 percent accuracy, followed by GPT-3.5 at 47 percent, and Llama-2 at 45 percent. The models showed significantly reduced accuracy on tasks with suspected hierarchical bias. For example, GPT-4's accuracy dropped to 33 percent on these tasks, compared to 86 percent on others. However, the models identified the nearest cardinal direction in most cases, reflecting their associative learning mechanism, thereby embodying human-like misconceptions. We discuss avenues for improving the spatial reasoning capabilities of LLMs.

cs.CL↗

arXiv2022

Estimating the potential shift from conventional public transport to flexible services based on smartcard transactions

We assume that urban travelers may prefer flexible modes of transportation over conventional public transport (PT) for making non-routine trips, and estimate the potential for such modal switch based on a database of 63 million smartcard records of PT boardings made in Israel during June 2019. The behavioral patterns of PT users are revealed by clustering their boarding records based on the location of the boarding stops and time of day, applying an extended DBSCAN algorithm. Our major findings are that (1) conventional home-work-home commuters are a minority and constitute less than 15% of the riders; (2) at least 30% of the PT trips do not belong to any cluster and can be classified as occasional; (3) The vast majority of users make both recurrent and occasional trips. A linear regression model provides a good estimate (R2 = 0.85) of the number of occasional boardings at a stop as a function of the total number of boardings, time of day, and land use composition around the location of trip origin. We conclude that the conventional PT may lose substantial urban ridership to the future flexible modes.

physics.soc-ph↗

arXiv2022

The nature of the on-street parking search

Parking occupancy in the area is defined by three major parameters - the rate of cars arrivals, the dwell time of already parked cars, and the willingness of drivers who are searching but yet did not find a vacant parking spot, to continue their search. We investigate a series of theoretical and numeric models, deterministic and stochastic, that describe parking dynamics in the area as dependent on these parameters, over the entire spectrum of the demand to supply ratio, focusing on the case when the demand is close to or above the supply. We demonstrate that a simple deterministic model provides a good analytical approximation for the major characteristics of the parking system - the average fraction of cars among the arriving that will find parking in the area, the average number of cars that cruise for parking, and average cruising time. Stochastic models make it possible to estimate the distributions of these characteristics as well as the parameters that are related to the variance of these distributions, like the fraction of the arriving cars that find parking in less than t minutes.

physics.soc-ph↗

arXiv2019

Establishing and assessing adaptive parking prices in a city: Algorithms, software and examples

We propose ParkSage, a set of spatially-explicit algorithms for establishing parking prices that guarantee a predetermined occupancy rate over a city, and for evaluating the achieved reduction in parking search time. We apply ParkSage for establishing overnight parking prices that guarantee 85% occupation in the Israeli city of Bat Yam. Pricing by street links ensures high parking availability and close to zero cruising everywhere in the city, but is inconvenient for drivers. Establishing prices by the large and heterogeneous city quarters results in local mismatch between demand and supply, the emergence of areas with fully occupied on-street parking and a long search time for the drivers whose destinations are in these areas. We demonstrate that pricing by the medium sized Transportation Analysis Zones, which is easy enough for drivers to comprehend and abide by, is sufficient for eliminating cruising. The software for establishing and assessing performance parking prices is based on the standard municipal GIS layers of streets and parking lots and is available for free download from https://www.researchgame.net/profile/Nir_Fulman

physics.soc-ph↗

arXiv2019

Approximation of Search Times for On-street Parking Based on Supply and Demand

We propose a method for approximating the probability p(τ, n) of searching for on-street parking longer than time τ from the start of a parking search near a given destination n, based on high-resolution maps of parking demand and supply in a city. We verify the method by comparing its outcomes to the estimates obtained with an agent-based model of on-street parking search. As a practical example, we construct maps of cruising time for the Israeli city of Bat Yam, and demonstrate that despite the low overall demand-to-supply ratio of 0.65, excessive demand in the city center results in parking searches of longer than 10 minutes. We discuss the application of the proposed approach for urban planning.

physics.soc-ph↗

arXiv2017

An algorithm and software for establishing heterogeneous parking prices

Parking prices in cities are uniform over large areas and do not reflect spatially heterogeneous parking supply and demand. Underpricing results in high parking occupancy in the subareas where the demand exceeds supply and long search for the vacant parking, whereas overpricing leads to low occupancy and hampered economic vitality. We present Nearest Pocket for Prices Algorithm (NPPA), a spatially explicit algorithm for establishing on-and off-street parking prices that guarantee a predetermined uniform level of occupation over the entire parking space. We apply NPPA for establishing heterogeneous parking prices that guarantee 90% parking occupancy in the Israeli city of Bat Yam.

physics.soc-ph↗