Energy optimization in autonomous driving using deep reinforced learning and stochastic system modelling
Akash Ladha
, Divya Priyadharshini Mohan
Electric vehicles, Fuel economy, Exhaust emissions, Battery state of charge, Deep reinforced learning
In the development of future low-emission vehicles, Machine Learning is playing a significant role, since manufacturers progressively hit constraints with existing technology. In addition to independent driving, new improvements in reinforcement learning are also quite good at handling complicated parameterisation challenges. Deep reinforced training is utilised in this research to derive efficient electric hybrid vehicle operating methods. A wide range of possible driving and traffic conditions should be predicted, so that fuel-efficient solutions may be achieved in order to achieve intelligent and adaptable processes. This study demonstrates a reinforced learning agent's capacity to learn almost optimum operational strategies without previous route knowledge and gives a large potential for more factors to be included in the optimization procedure. This paper includes (1) a deep learning context that will enable discovering virtually optimum operating strategies. (2) The use of stochastic driver models to increase public generalisation and prevent overfitting of the approach. (3) Inclusion of the optimization process of battery modelling with extra power restrictions. The results are simulated and comparison graphs are plotted for the derived model.
"Energy optimization in autonomous driving using deep reinforced learning and stochastic system modelling", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.6, Issue 9, page no.132 - 140, September-2021, Available :https://ijsdr.org/papers/IJSDR2109021.pdf
Volume 6
Issue 9,
September-2021
Pages : 132 - 140
Paper Reg. ID: IJSDR_193669
Published Paper Id: IJSDR2109021
Downloads: 000347181
Research Area: Engineering
Country: Kolkata, West Bengal, India
ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: IJSDR(IJ Publication) Janvi Wave
