Approaches and learning phases for a robotic adaptive deambulation task

The number of elderly people increases each year and their needs on technical aids represent a great challenge for robotics. This article presents a study about a mobility aid system integrating robotics and learning adaptive control. The system depends on using an auto-adaptive interface that improves the interpretation of the patient driving actions. The experimentation results show the advantage of such an approach for a deambulation task. We present two methods for an adaptive deambulation mode.

[1]  Kenneth M. Dawson-Howe,et al.  The application of robotics to a mobility aid for the elderly blind , 1998, Robotics Auton. Syst..

[2]  Kurt Hornik,et al.  Degree of Approximation Results for Feedforward Networks Approximating Unknown Mappings and Their Derivatives , 1994, Neural Computation.

[3]  Steven Dubowsky,et al.  PAMM - a robotic aid to the elderly for mobility assistance and monitoring: a "helping-hand" for the elderly , 2000, Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation. Symposia Proceedings (Cat. No.00CH37065).

[4]  Andrew R. Barron,et al.  Universal approximation bounds for superpositions of a sigmoidal function , 1993, IEEE Trans. Inf. Theory.

[5]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[6]  Patrick Henaff,et al.  Fuzzy and neural control for mobile robotics experimentations , 1997, Proceedings of 1st International Conference on Conventional and Knowledge Based Intelligent Electronic Systems. KES '97.

[7]  Fernando Matía,et al.  Increasing Intelligence in Autonomous Wheelchairs , 1998, J. Intell. Robotic Syst..

[8]  E. Colle,et al.  Assistance robotisée basée sur une coopération entre la personne handicapée et la machine , 2000 .