Using sensor-data collected by a meet rollator for deriving outdoor accessibility information concerning mobility impaired people

Abstract

Nowadays, not all pedestrian paths are accessible for mobility impaired people. Especially paths which consist of complex public spaces and old streets constitute problems. However, no reliable information is (quickly) available about problems that cause hindrance. The sensing rollator called meet rollator is equipped with several sensors that provide detailed data about a range of factors which affect the accessibility of the outdoors environment. A meet rollator gathers data that is supplementary, new and different. The goal of this thesis is to use sensor-data collected by a meet rollator to develop a geo-database that provides insight into the accessibility of pedestrian routes for mobility impaired people who need a wheelchair, rollator or mobility scooter. In order to achieve this goal, the following research question is defined: Does the current setting of the meet rollator provide insight into the accessibility of pedestrian routes for mobility impaired people? If not, how can the current setting be improved? This research has evaluated the available sensor (handheld Global Positioning System (GPS), Real-Time Kinematic (RTK) GPS and camera) data collected by a meet rollator. Literature study showed that in order to determine walking restrictions in the outdoor activity space of people with mobility impairments, the geometrical demands for movement can be taken into account. According to the Manual-book of Accessibility and an expert Job Haug, a good pedestrian path for people with a wheelchair, rollator and scoot mobile has to meet three geometrical demands for movement. This means that a pedestrian path forms an obstacle if it is narrower than 0.9 meter, steeper than 9.46° or if the height difference on the path is higher than 0.05 meter. The outcome, according to the assessment of the three available sensors (handheld Global Positioning System (GPS), Real-Time Kinematic (RTK) GPS and camera) of the meet rollator by using the knowledge of the Master of Science program Geomatics, is that with a handheld GPS no precise positioning can be reached. Therefore handheld GPS data is not useful for generating accessibility information. According to the knowledge the two sensors RTK GPS and camera are useful for generating accessibility information. The outcome, according to the implementation of the developed workflow for deriving outdoor accessibility information for a selected research area, is that some paths in the outdoor environment of Amsterdam require improvements to make them accessible for mobility impaired people. Furthermore through data processing, more insight is gained into the usefulness of RTK GPS and camera data. The results showed that the used sensors did not provide the needed precision. For the validation of the derived outdoor accessibility information a division is made between the validation of RTK GPS observations and camera observations. The BGT+ is proved suitable for providing more insight in and put semantics to the total observations obtained by implementing the workflows. The ground truth is proved suitable for validating the obtained accessibility information derived from RTK GPS and camera data. Finally, the results of an algorithm made by the Urban Modelling Group (UMG) of the University College Dublin are proved suitable for validating the obtained accessibility information derived from camera data. A geo-database that structures the validated accessibility information is created in order to give an overview of the accessible and non-accessible paths for the research area. The derived accessibility information obtained via camera data can best be stored by using polygons, to represent complex areas. The derived accessibility information obtained via RTK GPS data can best be stored by using line strings. In conclusion, this MSc thesis has demonstrated that the current setting of the meet rollator provides insight into the accessibility of pedestrian routes for mobility impaired people. However, the used sensors did not provide the needed precision. In order to improve the current setting of the meet rollator the usefulness of other sensors should be further investigated.