Observability and Sensor Placement Problem on Highway Segments: A Traffic Dynamics-Based Approach

Traffic congestion is a major problem on highways. To analyze this problem and then design different ways of reducing congestion, researchers need to observe traffic conditions on highways. Sensors are used to record and collect traffic data on highways. However, sensors must be placed efficiently to maximize the information collected and minimize monetary cost. This paper presents a novel approach for studying the observability problem on highway segments by utilizing linearized traffic dynamics about steady-state flows. First, we analyze the observability problem in terms of sensor placement and then present a method for comparing scenarios having different sensor placements along a highway. Different sensor placement scenarios are compared using the condition number of the observability matrix for the modeled system. Simulations are performed for various different numbers of highway cells, and then, generalized results are provided. We also discuss steps needed to extend this methodology to a general traffic network system.