Learning Traffic Incident Detectors
نویسندگان
چکیده
Automated traffic monitoring and traffic incident detection is becoming a reality thanks to expanding networks of traffic sensors placed on US highways. However, the current state-of-the-art incident detection systems employ algorithms that require significant manual tuning. In this work, we study machine learning traffic detection solutions that are intended to reduce the need for the time consuming setup. We show that combining a set of simple data traffic sensor data streams via classification methods is a promising way to obtain a detector with an acceptably low false-positive rate and high and fast recall. We build and test a number of SVM-detector solutions and their refinements and show that they can outperform the widely used baseline, the California 2 algorithm. The algorithms are tested on incident data obtained for a section of monitored highway in a metropolitan area.
منابع مشابه
Towards a Learning Traffic Incident Detection System
The state of the art in traffic incident detection is dominated by approaches that require significant manual tuning. Our hypothesis is that these time-consuming solutions can be sucessfuly eliminated with the help of machine learning methods and past traffic data collected nowadays on major highways. We show that combining the output of a set of simple, imperfectly tuned, “offthe-shelf” detect...
متن کاملLearning to Detect Adverse Traffic Events from Noisily Labeled Data
Many deployed traffic incident detection systems use algorithms that require significant manual tuning. We seek machine learning incident detection solutions that reduce the need for manual adjustments by taking advantage of massive databases of traffic sensor network measurements. First, we show that a rather straightforward supervised learner based on the SVM model outperforms a fixed detecti...
متن کاملAn Integrated Knowledge-base/simulation System for Real- Time Incident Management
This paper presents a real-time incident management system that integrates the knowledge base with a microscopic traffic simulation model. The system consists of five modules: input module, knowledge-based module, prediction module, simulation module, and output module. The knowledge base is used to inventory the operational experience and traffic impact information associated with all recorded...
متن کاملIncident Detection with Traffic sensors on urban highways
by ryan fries, ph.d., p.e., mashrur a. chowdhury, ph.d., p.e., yongchang ma, ph.d., and anne e. dunning, ph.d. inTroducTion Traffic congestion is an increasingly significant problem on America’s highways, wasting time and fuel while increasing emissions.1 Most American roadways (approximately 60 percent) are already congested, and drivers spend more time traveling than ever before.2 Reducing co...
متن کاملDevelopment of Unequally-Spaced Traffic Measurement Prediction Model
This paper presents the development of unequally-spaced or irregularly observed traffic measurement prediction model. Traffic measurements from detectors are usually observed or assumed to be equally-spaced (e.g., every 30 seconds and 1 minute) for transportation research. However, some traffic measurements may not be observed or assumed to be equally-spaced especially in arterial street and IT...
متن کامل