Title

Embedded Sensing of Transportation Networks

Abstract

Optimizing planning and operations of existing transportation networks requires an accurate measurement of its state. Some common types of measurement systems are loop detectors and video cameras. These approaches do not provide a complete characterization of the state of the network. Moreover, they are expensive to install and operate. In this talk we describe an alternative measurement platform: road embedded wireless sensor nodes. Combined with new algorithms we have designed, such nodes are capable of providing travel times, queue lengths, vehicle classes and vehicle loads. I will present various algorithms for estimating these quantities and if time permits, a provably optimal decentralized control algorithm for coordinating traffic signals based on the novel measurements. Some of the key advantages of the technology we present are its high reliability, novel measurement capabilities and low installation and operating cost.