User intent in a shared control framework for pedestrian mobility aids

Abstract

This paper presents a novel approach to infer navigational intent of the user of a walker, based on measuring forces and moments applied to the walker's handles. While there are many types of "intent" that could be inferred for a given user action, the experiments conducted here focused on the determining user's navigational intent, i.e. their desired heading. Our experiments used two 6-DOF force/ moment sensors on the walker's handles and a digital motion capture system to correlate applied force with actual motion. Preliminary results revealed that the intent to turn, represented by changes in the heading angle, highly correlates with the overall turning moment around the vertical axis as well as the side forces applied by the user. Other force/moment components reveal additional information, such as support needs. The inferred user intent will be incorporated into a passive shared steering control system for the walker.

User Intent Communication in Robot-Assisted Shopping for the Blind

“…Under the first scenario, the user does not communicate with the robot explicitly. The robot attempts to infer or predict user intent from its own observations (Wasson et al, 2003;Demeester et al, 2006). Under the second scenario, the user communicates intent to the robot with body gestures (Morency et al, 2007).…”

Basic walker-assisted gait characteristics derived from forces and moments exerted on the walker's handles: Results on normal subjects

“…The Vicon motion capture system was chosen as a reference in the experiments, since the experiments also aimed at capturing the trajectory of the walker as well as the user's forces and moments in different scenarios in attempt to understand the User's Navigational Intent (see Ref. [3]). The global coordinate system of Vicon is aligned with the local coordinate system of the force-moment measurement system, which is tied to the walker's frame, at the beginning of every experiment.…”

“…This paper builds on previous research performed at the University of Virginia in using instrumented wheeled walkers to develop a human/machine shared-control system that assists users by increasing the safety and speed of their daily travel [3,12,13]. Here we present the use of a walker instrumented with force-moment sensors to passively derive basic gait characteristics.…”

“…In several research projects walkers have been instrumented for control purposes, such as the inference of the intent of the walker's user to implement control [3]. However, few researchers have attempted to instrument a walker to assess the functionality of the walker-dependent user.…”

Mode Switch Assistance To Maximize Human Intent Disambiguation

“…Any assistive autonomy system typically needs an idea of what it is the human is trying to do-either by explicit indication from the user of the task or goal [6], or by inferring the human's intent from their control signals or sensor data [22,25]. The question of intent inference thus is key.…”