@inproceedings {pub3181pub3211,
	title = {Automatic Energy Management Controller Design for Hybrid Electric Vehicles},
	author = {Tobias Rodemann AND Lars Gr{\"a}ning AND Ken Nishikawa},
	year = {2016},
	month = {December},
	abstract = {Due to strict CO2  emission limits, the optimal design of controllers for hybrid cars is an increasingly important topic. Most current approaches use engineers{\textquoteright} knowledge to develop controllers. In this work we evaluate how simple control rules can automatically be extracted from optimal controls computed by Dynamic Programming (DP). We compare artificial neural networks and decision trees in terms of performance (fuel consumption), stability, robustness, and interpretability. We also investigate how the derived controllers depend on the specific drive cycles used for generating optimal control. Our findings indicate that automatically derived controllers can provide performance within 1-2\% above optimal fuel consumption, but we also see a large variety in performance and controller structures for different drive cycles. },
	publisher = {IEEE},
	booktitle = { Computational Intelligence (SSCI), 2016 IEEE Symposium Series on},
	city = {Athens},
	pages = {1-8}
}