Co-Investigators:
Annotation:
Most analyzes in the field of transportation planning require knowledge of the origin and destination of a trip, often aggregated into flows and described by an Origin-Destination (OD) matrix. Considering that it is substantial to have proper OD matrices when simulating traffic, many approaches have been proposed to solve the OD estimation problem. Direct measurements/interviews or surveys are difficult and often costly, hence aggregate methods that use traffic counts and other available information are used to obtain reasonable OD estimates. The purpose of this project is to describe and estimate OD matrices for vehicles, bicycles, and pedestrians using traffic counts, travel times, and movements within intersections. First, we will conduct a comprehensive analysis of current approaches. Then, we will identify the road sections (links) to be included in the network and define those as origin and destination. For vehicles and bicycles, the solution should generate sets of feasible flows and routes between origins and destinations with given turning rates and traffic counts constraints, and iteratively update them until a stopping criterion, e.g., fitness to real travel times from Google Maps API. For pedestrians, OD flows should be based on the shortest routes, constrained by the movements within intersections, and influenced by the type and size of buildings along the route (that can be collected from OpenStreetMaps).
Department:
Year:
2023 - 2024
Program:
Studentská grantová soutěž ČVUT - SGS23/129/OHK2/2T/16
