The ability of road infrastructure in New Brunswick to support Lane Keep Assist technology
| dc.contributor.advisor | Hildebrand, Eric | |
| dc.contributor.author | Stewart, Michael | |
| dc.date.accessioned | 2023-09-21T18:01:05Z | |
| dc.date.available | 2023-09-21T18:01:05Z | |
| dc.date.issued | 2021-11 | |
| dc.description.abstract | This study expands our understanding of how well lane keep assist (LKA) technologies perform under varying conditions of pavement markings and lane edge treatments. Freeways, Arterials and Collectors that typically benefit from higher maintenance standards for pavement markings were able to support the LKA system throughout 95% to 100% of their length. These findings do not include weather events where snow/ice accumulation resulted in partial to complete covering of both the centre line marking, and edge of roadway. The compliance percentages include wet & dry surface conditions, varied rain/snow intensities and visibility hinderances caused by fog. These road segments possessed contrast ratios of 2.23 or greater along with retroreflectivity values of 60.7 mcd/lx/m2 or greater. The impact of traversing shadows or the provision of streetlight illumination had little to no effect on the machine vision system (MVS) and consequently the LKA performance. Road segments with edge characteristics other than a painted edge line (i.e., crushed rock, gravel, grass) had mean LKA engagement ranging from only 5.7% to 61.0% and were greatly impacted by lighting conditions, environmental conditions, and deteriorated edge (raveling) of asphalt commonly found on Local highways. These road segments typically possessed contrast ratios and retroreflectivity values less than the recommended guidelines of 1.6, and 34 mcd/lx/m2 for daytime and nighttime dry operations. Road segments in New Brunswick should possess contrast ratios greater than 1.9, and retroreflectivity values of 65 mcd/ls/m2 or greater to ensure LKA engagement of 90% or higher. | |
| dc.description.copyright | ©Michael J. Stewart, 2021 | |
| dc.format.extent | x, 127 | |
| dc.format.medium | electronic | |
| dc.identifier.oclc | (OCoLC)1419001457 | en |
| dc.identifier.other | Thesis 10990 | en |
| dc.identifier.uri | https://unbscholar.lib.unb.ca/handle/1882/37443 | |
| dc.language.iso | en | |
| dc.publisher | University of New Brunswick | |
| dc.rights | http://purl.org/coar/access_right/c_abf2 | |
| dc.subject.discipline | Civil Engineering | |
| dc.subject.lcsh | Artificial intelligence. | en |
| dc.subject.lcsh | Road markings--New Brunswick. | en |
| dc.subject.lcsh | Automobiles--Electronic equipment--New Brunswick. | en |
| dc.subject.lcsh | Automobiles--New Brunswick--Automatic control. | en |
| dc.subject.lcsh | New Brunswick. | en |
| dc.title | The ability of road infrastructure in New Brunswick to support Lane Keep Assist technology | |
| dc.type | master thesis | |
| oaire.license.condition | other | |
| thesis.degree.discipline | Civil Engineering | |
| thesis.degree.grantor | University of New Brunswick | |
| thesis.degree.level | masters | |
| thesis.degree.name | M.Sc.E. |