Surface electromyography (sEMG) has been widely used as the control command for assistive devices because the activation of sEMG signals precedes the actual human movement. Such time advantage allows the minimization of mechanical resistance felt between the user's movement and the assistive device. In this study, we investigated the feasibility of identifying five environments (flatland, slope up, slope down, stair up, stair down) using four channels of sEMG extracted from the vastus medialis (VM), hamstring (HAM), tibialis anterior (TA), and gastrocnemius (GAS). We selected the four muscles considering the wearability of the user. We collected the sEMG data for 15 steps in each environment and compared the integrated electromyography(IEMG) results. Results of the IEMG show that the five environments can be differentiated using basic pattern
recognition methods (i.e. Fuzzy).