Fuzzy expert system for determining the human gait phase

Abstract

This paper describes fuzzy logic expert system (FES) uses data about the Zero Moment Point (ZMP) position for automatically determining the gait phases. Simplified human body model is introduced for computation of ZMP position which is the input to the FES. Information about the type of gait phase is the output of FES. The designed fuzzy-rules based system was tested to evaluate the gait of twenty-two volunteers. Results show that the proposed technique for determining the gait phases is suitable. In the case of identification of double supports, the FES was successful in 73%. In the case of identification of single supports, the FES was successful in 64%. The method for studying ZMP in combination with fuzzy logic could contribute to the design of new control systems of prosthesis. Also, the technique can be used for specific rehabilitation examinations and the diagnosis of disorders of the musculoskeletal system.