Loading...

Design and Implementation of a Control System based on Human Intention Detection by Measuring the Interaction Force for Lower-Limb Exoskeleton

Rashidi, Amir Reza | 2016

605 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 48306 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Vossoughi, Gholamreza; Zohoor, Hassan
  7. Abstract:
  8. The control approaches in Exoskeletons are different from Autonomous robot. Because the Exoskeletons are in connection and interaction with the user. On the other hand they can use the wearer’s abilities such as intelligence, sensory organs, self-control and etc. Therefore, one of the most important parts of exoskeletons control algorithms is its interaction with the user. This part should recognize the goal and purpose of the wearer such as stopping, walking start, change in stepping and direction and etc. properly. Utilizing these algorithms properly, one can reduce the muscle activities and energy consumption up to 20 percent [1]. This investigations demonstrate the importance of the algorithms to make the robot helping the wearer properly. In this research, a control strategy for exoskeletons will be presented which is based on the recognition of human intention. The algorithm will be investigated and designed for power augmentation of healthy people and includes all 3 control levels of exoskeletons. For this purpose, the stop and start phase and single support (swing) and double support phase of the walking has been investigated. For each sub-phase of walking, a mid-level controller has been developed based on the interaction force of the robot and human. In the high level of control, a fuzzy state machine has been developed for recognition of human intention and transmitting time of the sub-phases
  9. Keywords:
  10. Fuzzy State Machine ; Interactive Control System ; Exoskeleton ; Human Intention Recognition ; Power Augmentation ; Walking Pattern Recognition

 Digital Object List

 Bookmark