Using Support Vector Machines for Lane-Change Detection

Abstract

Driving is a complex task that requires constant attention, and intelligent transportation systems that support drivers in this task must continually infer driver intentions to produce reasonable, safe responses. In this paper we describe a technique for inferring driver intentions, specifically the intention to change lanes, using support vector machines (SVMs). The technique was applied to experimental data from an instrumented vehicle that included both behavioral data and environmental data. Comparing these results to recent results using a novel “mind-tracking” technique, we found that SVMs outperformed earlier algorithms and proved especially effective in early detection of driver lane changes.