Title:
Traffic light prediction using connected vehicles
Traffic light prediction using connected vehicles
| dc.contributor.advisor | Bras, Berdinus A. | |
| dc.contributor.advisor | Jiao, Jianxin (Roger) | |
| dc.contributor.advisor | Simmons, Richard | |
| dc.contributor.author | Mcarthur, Christopher Thomas | |
| dc.contributor.department | Mechanical Engineering | |
| dc.date.accessioned | 2020-05-20T16:57:43Z | |
| dc.date.available | 2020-05-20T16:57:43Z | |
| dc.date.created | 2019-05 | |
| dc.date.issued | 2019-04-24 | |
| dc.date.submitted | May 2019 | |
| dc.date.updated | 2020-05-20T16:57:43Z | |
| dc.description.abstract | Automotive companies have focused on reducing emissions of vehicles through their design. However, there is opportunities for larger emissions reductions through coaching the driver on how improve fuel economy. Driver behavior has an impact on the fuel economy of the vehicle. Gains between 3 and 20 percent can be observed through altering how the driver operates the vehicle. How can the driver be coached in new ways to improve fuel economy? Many of the traditional approaches focus on having the driver using the throttle and brake pedals less aggressively. The approach in this paper implements coaching where the driver is advised based on what is happening in the environment around the vehicle. The environmental events that are being coached on is the timings of traffic lights. A prototype application was constructed that implemented all of these techniques. The system was implemented in real time using an android app. The system took information from the traffic light information files to successfully inform the driver on the environment that they were driving through. The system has been implemented through a prototype application and the results are as follows. Traffic light prediction was successful at predicting 2 cycle fixed time traffic lights. These fixed time traffic lights account for the most common traffic lights in the U.S. The traffic light prediction algorithm and work in that field is promising. The leader-follower traffic light prediction coaching shows fuel consumption reduction by as much as 34% in extreme cases. An average fuel consumption improvement of 18.7% is observed for all the drivers tested. | |
| dc.description.degree | M.S. | |
| dc.format.mimetype | application/pdf | |
| dc.identifier.uri | http://hdl.handle.net/1853/62694 | |
| dc.language.iso | en_US | |
| dc.publisher | Georgia Institute of Technology | |
| dc.subject | Driver coaching | |
| dc.subject | Traffic light prediction | |
| dc.subject | Autonomous vehicles | |
| dc.title | Traffic light prediction using connected vehicles | |
| dc.type | Text | |
| dc.type.genre | Thesis | |
| dspace.entity.type | Publication | |
| local.contributor.advisor | Jiao, Jianxin (Roger) | |
| local.contributor.advisor | Bras, Berdinus A. | |
| local.contributor.corporatename | George W. Woodruff School of Mechanical Engineering | |
| local.contributor.corporatename | College of Engineering | |
| relation.isAdvisorOfPublication | 48ca5e07-55f0-41b0-a2f4-9c033612965f | |
| relation.isAdvisorOfPublication | 9c522ea2-cfd8-4ff0-ac9c-b62b07f7c32a | |
| relation.isOrgUnitOfPublication | c01ff908-c25f-439b-bf10-a074ed886bb7 | |
| relation.isOrgUnitOfPublication | 7c022d60-21d5-497c-b552-95e489a06569 | |
| thesis.degree.level | Masters |