German H. Flores

Ariadne's thread: Robust turn detection for path back-tracing using the iPhone

Most systems for pedestrian localization and self-tracking aim to measure the precise position of the walker and match it against a map of the environment. In some cases, a simpler topological description of the path taken may suffice. This is the case for the system described in this paper, which is designed to help a blind person re-trace the route taken inside a building and to walk safely back to the starting point. We present two turn detection algorithms based on hidden Markov models (HMM), which process inertial data collected by an iPhone kept in the walkers front pocket, without the need for a map of the environment. Quantitative results show the robustness of the proposed turn detectors even in the case of drift in the measurements and noticeable body sway during gait.

Experiments with a Public Transit Assistant for Blind Passengers

Public transportation is key to independence for many blind persons. Unfortunately, in spite of recent accessibility progress, use of public transportation remains challenging without sight. In this contribution, we describe a system that provides enhanced travel-related information access to a blind bus passenger. Users of this system can select a specific bus line and desired destination on a regular Android smartphone or tablet; are notified when the bus arrives; once in the bus, they are informed of its progress during the route; and are given ample advance notice when the bus is approaching their destination. This system was tested with four blind participants in realistic conditions.

Easy Return: An App for Indoor Backtracking Assistance

We present a system that, implemented as an iPhone app controllable from an Apple Watch, can help a blind person backtrack a route taken in a building. This system requires no maps of the building or environment modifications. While traversing a path from a starting location to a destination, the system builds and records a path representation in terms of a sequence of turns and of step counts between turns. If the user wants to backtrack the same path, the system can provide assistance by tracking the users location in the recorded path, and producing directional information in speech form about the next turns and step counts to follow. The system was tested with six blind participants in a controlled indoor experiment.

Transit Information Access for Persons with Visual or Cognitive Impairments

We are developing a location-based information delivery system to facilitate efficient and safe use of public transportation by people who have visual or cognitive impairments. This system comprises Wi-Fi beacons (access points) that are placed at bus stations and inside bus vehicles. Users of this system receive information on their cell phone, without the need for GPS or for Internet connectivity. The system allows one to receive information about an upcoming bus at a bus stop and to select a specific bus line. Once the desired bus arrives, the system automatically connects to the access point on the bus vehicle and remains connected while the user is riding the bus. The user can specify a desired bus stop, and the system informs the user (by a speech message) with enough advance notice when the bus is approaching the stop.

WeAllWalk: An Annotated Dataset of Inertial Sensor Time Series from Blind Walkers

We introduce WeAllWalk, a dataset of inertial sensor time series collected from blind and sighted walkers using a long cane or a guide dog. Ten blind volunteers (seven using a long cane, one using a guide dog, and two alternating use of a long cane and of a guide dog) as well as five sighted volunteers contributed to the data collection. The participants walked through fairly long and complex indoor routes that included obstacles to be avoided and doors to be opened. Inertial data were recorded by two iPhone 6s carried by our participants in their pockets and carefully annotated. Ground-truth heel strike times were measured by two small inertial sensor units clipped to the participants’ shoes. We also present an in-depth comparative analysis of various step counting and turn detection algorithms as tested on WeAllWalk. This analysis reveals interesting differences between the achievable accuracy of step and turn detection across different communities of sighted and blind walkers.