Home // ADAPTIVE 2018, The Tenth International Conference on Adaptive and Self-Adaptive Systems and Applications // View article
Authors:
Karl-Falco Storm
Daniel Eckardt
Meng Zhang
Jörg Grieser
Michael Prilla
Rausch Andreas
Keywords: pattern recognition; long-term prediction; driver condition monitoring; primitive driving behavior; model predictive control.
Abstract:
This paper evaluates the prediction accuracy of indeterminis-tic environments. The exhaust aftertreatment for vehicles is used as a sample scenario, whose efficiency should be en-hanced using pattern recognition techniques. It determines a control strategy to minimize exhaust emissions—whose vol-ume, composition and temperature depends on the load and speed of the combustion engine. Since the engine being con-trolled by the accelerator pedal, the driving behavior needs to be predicted for adequate horizons. The new approach is simulated on the basis of driving data at different traffic sce-narios, including urban, overland and motorway road types. The recorded driving behavior is examined location-based by transferring it into a dynamical number of primitive driving behavior classes. This way, traffic scenarios can be distin-guished by using a relatively small set of data. Furthermore, the driving behavior does not have to be labeled, since infor-mation about it occurring is not required. In context with the task of vehicle control, possible changes in driving behavior due to a higher stress level have already been proven. Follow-ing this finding, driving behavior prediction is investigated in consideration of the driver’s condition. In the end, a bench-mark is carried out to compare existing prediction methods of location-based pattern recognition. After presenting the find-ings, an outlook for possible future research is given.
Pages: 84 to 92
Copyright: Copyright (c) IARIA, 2018
Publication date: February 18, 2018
Published in: conference
ISSN: 2308-4146
ISBN: 978-1-61208-610-1
Location: Barcelona, Spain
Dates: from February 18, 2018 to February 22, 2018