Autonomous, vision-based, pivoting wheelchair with obstacle detection capability

  • Authors:
  • Guillermo Del Castillo Del Riego;Steven B. Skaar

  • Affiliations:
  • -;-

  • Venue:
  • Autonomous, vision-based, pivoting wheelchair with obstacle detection capability
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

An advanced prototype computer controlled power wheelchair or CPWNS has been developed to provide autonomy for highly disabled users, whose mix of disabilities makes it difficult or impossible to control their own power chairs in their homes. The working paradigm is “teach and repeat” a mode of control for typical industrial holonomic robots. The industrial robot-control system stores internal joint poses that have been selected by a human operator and recalls them in sequence to execute each recurring task. Sequential pose replication for the “nonholonomic” wheelchair, while entailing additional, computationally intensive, considerations of estimation, computer-vision observations, collision-avoidance, and control, can be equally effective. Path segments that connect wheelchair positions/orientations of interest within the home or office are first taught by pushing the chair through the path that is to be recalled later. The teaching event occurs with the estimator “turned on”. Ultrasound sensors, which during subsequent autonomous tracking will be used to avoid obstacles, also are active during teaching. Based upon post-processed data collected during this teaching event, elaborate trajectories—which may involve multiple direction changes, pivoting and so on, depending upon the requirements of the typically restricted spaces within which the chair must operate—will later be called upon by the disabled rider. The postprocessor also assigns an ultrasound profile to the sequence of poses. Use of this profile during tracking obviates most of the inherent problems of using ultrasound to avoid obstacles while retaining the ability to near solid objects where required by the environment and trajectory objectives. Likewise, use of the same wall-mounted visual cues for reference/estimation during both the teaching and tracking events eliminates the need for globally accurate prior mapping of the cue locations. The vehicle has been tested extensively and successfully with disabled veterans at the Edward J. Hines, Jr. VA Hospital in Hines, IL.