View More
View Less
System Message
An unknown error has occurred and your request could not be completed. Please contact support.
Wait Listed
Personal Calendar
Conference Event
Times for this session to be announced soon
Conflict Found
This session is already scheduled at another time. Would you like to...
Please enter a maximum of {0} characters.
{0} remaining of {1} character maximum.
Please enter a maximum of {0} words.
{0} remaining of {1} word maximum.
must be 50 characters or less.
must be 40 characters or less.
Session Summary
We were unable to load the map image.
This has not yet been assigned to a map.
Search Catalog
Replies ()
New Post
Microblog Thread
Post Reply
Your session timed out.
This web page is not optimized for viewing on a mobile device. Visit this site in a desktop browser to access the full set of features.
LiveWorx 2017

A Smarter Car Through Machine Learning

Session Description

Machine learning enables us to extract and utilize information that lies within the data generated by millions of connected devices each day. It has the potential to make everyday devices smarter, and our cars are no exception. But what does it take to transform an ordinary car into a smart car? We have developed a prototype of a Smart Car that recognizes its driver by analyzing data that it already collects from its internal sensors. We collected and decoded time-series sensor data such as gear position, steering wheel position, and RPMs, during several trips with four different drivers. We trained a model that takes this data as input, associates each driver with patterns in the data, determines the probability that each of the four drivers is the one currently driving the vehicle, and reports the most likely driver. In our tests, the Smart Car model successfully identified the driver with 92% accuracy. In this session, we will discuss some of the challenges involved in developing a connected car. We will give an overview of general machine learning concepts and the specific techniques used in our prototype, including feature engineering, feature selection, time series prediction, and categorical prediction. Finally, we will demonstrate the Smart Car’s ability to identify drivers based on their driving patterns.

Session Presenter
Additional Information
Analytics & Big Data
Analytics can add tremendous value to IoT applications by transforming existing data into decisions, predictions, descriptions, and alerts.
Machine learning can compensate for missing, unreliable, or difficult-to-measure information by making inferences from other information that is available.
Analytics can improve efficiency. It removes the need for time-consuming manual inspection of large amounts of data by human experts, and allows decisions to be made automatically and in real time.
Breakout Session
45 minutes
Session Schedule