Piyawan Kasemsuppakorn

Personalised routing for wheelchair navigation

Wheelchair users are confronted with many outdoor obstacles in everyday travel that can make the task arduous and unsafe, especially in unfamiliar environments. However, if wheelchair users are provided with routes that can safely, accurately and efficiently guide them to their destinations, then the difficulty and danger associated with traveling might be significantly reduced. In this article, we present a routing method suitable for wheelchair users by taking into account sidewalk obstacles such as slope, steps, sidewalk condition and sidewalk traffic as are preferred by users. Our method calculates impedance scores for each sidewalk segment, which are then used to determine an optimal route between any given pair of origin-destination addresses in a sidewalk network. We first discuss sidewalk network parameters and a spatial database suitable for wheelchair navigation. Then we discuss our personalised routing approach. We also describe three methods to weight the impedance level of each sidewalk segment and compare the effectiveness of each method through experimentation using the sidewalk network for the University of Pittsburgh campus area.

Pedestrian network map generation approaches and recommendation

With the advanced capabilities of mobile devices and the success of car navigation systems, interest in pedestrian navigation systems is on the rise. A critical component of any navigation system is a map database, which represents a network (e.g., road networks for car navigation) and supports key functionality such as map display, geocoding, and routing. Road networks, mainly due to the popularity of car navigation systems, are well defined and publicly available. However, in pedestrian navigation systems, as well as other applications including urban planning and physical activity studies, road networks do not adequately represent the paths that pedestrians usually travel. Currently, there is a void in literatures discussing the challenges, methods, and best practices for pedestrian network map generation. This coupled with the increased demand for pedestrian networks is the prime motivation for development of new approaches and algorithms to automatically generating pedestrian networks. Three approaches, network buffering, using existing road networks, collaborative mapping, using Global Positioning System (GPS) traces collected by volunteers, and image processing, using high-resolution satellite and laser imageries, were implemented and evaluated with a pedestrian network baseline as a ground truth. The results of the experiments indicate that these three approaches, while differing in complexity and outcome, are viable for automatic pedestrian network map generation. The recommendation of a suitable approach for generating pedestrian networks for a given set of sources and requirements is provided.

Design Considerations for a Personalized Wheelchair Navigation System

Individuals with mobility impairments such as wheelchair users are often at a disadvantage when traveling to a new place, as their mobility can be easily affected by environmental barriers, and as such, even short trips can be difficult and perhaps impossible. We envision a personalized wheelchair navigation system based on a PDA equipped with wireless Internet access and GPS that can provide adaptive navigation support to wheelchair users in any geographic environment. Requirements, architectures and components of such a system are described in this paper.

Pedestrian network data collection through location-based social networks

The increasing capabilities of mobile devices and their mobile positioning technologies have shown great promise in location-enabled applications such as navigation systems. One of the essential components of a navigation system is a spatial database as it provides the base data to perform navigation and routing functions, among others. With the popularity of vehicle navigation systems, road network databases are now well developed and well suited for vehicles traveling on roads. However, road networks are not suitable for pedestrian navigation systems as pedestrians do not walk along the middle of street lanes and are not constrained by the boundaries of the road. Consequently, pedestrian network data is needed in location-enabled applications for pedestrians and other applications including transportation planning and physical activities study. Due to the lack of availability of pedestrian network data, new approaches for acquiring and maintaining pedestrian network data that are efficient and cost effective are needed. This paper presents a new approach that is based on location-based social networking for collecting pedestrian network data. The proposed approach stems from the concept of collaborative mapping where pedestrian network data can be collected by members of a social network recording GPS trajectories. A prototype based on a framework called Social Navigation Network (SoNavNet) for sharing and recommending navigation related information is also discussed.

Uncertainty in personal navigation services

The demand for navigation assistance and advances in several technologies has been paving the way for Personal Navigation Services (PerNavs). As users increasingly rely on PerNavs for navigation assistance, they gain a better understanding of what PerNavs can offer and how they operate. This trend, consequently, will increase the demand for PerNav that can provide high quality solutions. While there have been studies addressing uncertainties associated with selected individual navigation modules, there is a void in the literature addressing the overall uncertainty in PerNavs. In this paper, we discuss uncertainty in PerNavs by analyzing uncertainties associated with each of its modules and how they propagate and impact other modules. A Bayesian network is presented as one possible model to manage (by developers) and communicate (to users) uncertainty in PerNavs.

Understanding route choices for wheelchair navigation

Abstract Purpose: To validate a personalized routing technique with wheelchair users, understand their route choices and acquire their feedback on the necessity of wheelchair navigation and the importance of personalized routes. Method: A routing technique using a weighting method, called Absolute Restriction Method (ARM), was employed to compute personalized routes based on users’ routing preferences. The evaluation involves five manual wheelchair users. The study protocol consists of three sessions: pre-activity, activity and post-activity sessions. The evaluation included a comparison between personalized routes and shortest feasible routes, in terms of route characteristics and users’ ratings of important parameters. Results: Subjects travelled a 14.64% longer distance along the personalized routes than the shortest feasible routes. However, all personalized routes had better path quality (slope and surface condition) than the shortest feasible routes. Four out of five subjects rated the parameters they deemed most important higher for the personalized route than for the shortest feasible route. Conclusions: The study confirmed that the shortest route criterion is not always suitable for individuals with mobility impairments. Personalized routes that take into account individual characteristics, route preferences and environmental characteristics are a promising solution to lessen the difficulties that manual wheelchair users face when navigating unfamiliar environments. Implications for Rehabilitation Wheelchair users indicate the importance of personalized routes for individuals with mobility impairments. In regard to evaluation results, although subjects travelled 14.64% more distances in average along the personalized routes than the shortest feasible routes, they rated the personalized routes better path quality and less effort to travel.

Pedestrian Network Extraction from Fused Aerial Imagery (Orthoimages) and Laser Imagery (Lidar)

A pedestrian network is a topological map that contains the geometric relationship between pedestrian path segments (e.g., sidewalk, crosswalk, footpath), which is needed in a variety of applications, such as pedestrian navigation services. However, current pedestrian networks are not widely available. In an effort to provide an automatic means for creating pedestrian networks, this paper presents a methodology for extracting pedestrian network from aerial and laser images. The methodology consists of data preparation and four steps: object filtering, pedestrian path region extraction, pedestrian network construction, and raster to vector conversion. An experiment, using ten images, was conducted to evaluate the performance of the methodology. Evaluation results indicate that the methodology can extract sidewalk, crosswalk, footpath, and building entrances; it collects pedestrian networks with 61 percent geometrical completeness, 67.35 percent geometrical correctness, 71 percent topological completeness and 51.38 percent topological correctness.