Loading...

Design and Implementatioan of a Locomotion mode Recognition Algorithm for Powered Lower-Limb Prosthesis

Shahmoradi, Sina | 2017

614 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 49622 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Bagheri Shouraki, Saeed
  7. Abstract:
  8. Control of powered lower limb prostheses has a locomotion mode- ependent structure which demands a pattern recognizer that can classify the current locomotion mode and also detect transitions between them in an appropriate time. In the way to achieve this goal, this project presents a locomotion mode recognition system to classify daily locomotion modes consist of level- walking, stair climbing, slope walking, standing and sitting using low-cost mechanical sensors. Since these signals have a quasi-periodic nature, using sequential pattern recognition tools, such as Hidden Markov Model(HMM) improves the recognition performance,because they use sequences of information to make a decision. On the other hand, simplicity in fuzzy linguistic description of patterns makes training and classification procedures easier. Also the imprecise nature of fuzzy modelling provides robustness to noisy signals from mechanical sensors which in turn raises hopes to improve the recognition performance. According to the mentioned reasons, the novel fuzzy sequential pattern recognizer named Fuzzy Elastic Matching Machine has been used as a recognizer which models the daily locomotion modes with patterns consist of some finite states and features of each state are described with a fuzzy vector. Dynamic locomotion modes have been modeled with four-state FEMMs. Sitting mode consisted of stand to sit, sitting and sit to stand states and static standing and dynamic standing constituted the states of standing mode. Generating database for training the fuzzy describing vectors and testing the performance of proposed system have been accomplished with the measured data from three IMUs and two FSRs mounted on the leg of four able-bodied subjects. FEMMs have been trainf with segmented gait cycles. Recognition performance has been evaluated for userspecific and user-independent classifications and the results are compared with HMM and LDA methods. Overall accuracy of 95.47% for user-specific classification and 86.15% for user-independent classification have been achieved, which indicated higher recognition accuracy compared with HMM and LDA methods. Also the proposed system was able to predict transitions between locomotion modes in an appropriate time
  9. Keywords:
  10. Fuzzy Elastic Matching Machine ; Hidden Markov Model ; Lower Limb Exoskeleton ; Locomotion Control ; Locomotion Mode Recognition ; Mechanical Sensors

 Digital Object List

 Bookmark

No TOC