cqu_6781+ATTACHMENT01+ATTACHMENT01.3.pdf (287.5 kB)

Smart driving : a new approach to meeting driver needs

Download (287.5 kB)
conference contribution
posted on 06.12.2017, 00:00 by G Shafiullah, Colin Cole, A B M Shawkat Ali
The use of machine learning algorithms in different automated applications is increasing rapidly. The effectiveness of algorithms performances helps the user to operate their machine accurately and on time. Road sign classification is a very common type of problem for an automated driving support system. In this research, road speeding measure and sign identification is conducted using four popular machine learning algorithms to develop a smart driving system. This system informs forward-looking decision making and the initiation of suitable actions to prevent any future disastrous events. The robustness of the classification algorithms is examined for classification accuracy through 10-fold cross validation and confusion matrix. Experimental results proofs that the accuracy of Support Vector Machine (SVM) and Neural Network (NN) is almost 100% and it is very promising compared to the earlier research performance. However, in terms of computational complexity NN is a slower classifier. Therefore, the experimental results suggest that SVM can make an effective interpretation and point out the ability of design of a new intelligent speed control system.

History

Start Page

1

End Page

7

Number of Pages

7

Start Date

01/01/2010

Location

Dhaka, Bangladesh

Publisher

IEOM

Place of Publication

Dhaka, Bangladesh

Peer Reviewed

Yes

Open Access

No

External Author Affiliations

East West University (Dhaka, Bangladesh); Faculty of Arts, Business, Informatics and Education; Faculty of Sciences, Engineering and Health; Institute for Resource Industries and Sustainability (IRIS);

Era Eligible

Yes

Name of Conference

International Conference on Industrial Engineering and Operations Management

Exports

Logo branding

Exports