An Improved Nonstationary Fuzzy System Approach versus Type-2 Fuzzy System for the Lifting Motion Control with Human-in-the-Loop Simulation

Abstract

People working in the fields of robotics, animation, computer graphics and computer vision, and biomechanics, find it difficult to conduct human motion simulations, such as lifting, walking and running. This is because it is difficult to predict all of the motion strategies in a variety of situations. The human lifting motion is hard work; at the same time, lifting is the most demanded robotic motion. In this paper, a fuzzy integral based nonstationary fuzzy inference system is proposed to control a five-segment human model for the human lifting motion with human-in-the-loop simulation. This system was designed to reduce the computational complexity of nonstationary fuzzy systems. Hence, this paper contributes to the literature in two ways: as an improved nonstationary fuzzy system and as a human lifting motion simulation framework. A fuzzy integral algorithm between the fuzzification and defuzzification stages has been used to aggregate the inference outputs of the nonstationary fuzzy controller. The fuzzy integral algorithm uses fuzzy values obtained during the fuzzification stage as the attribute values and the fuzzy values obtained by a one-loop quantum particle swarm optimization algorithm as the importance values. The computational complexity in the nonstationary fuzzy systems and type-2 fuzzy systems can be reduced between 25 to 60 percent with the improved nonstationary fuzzy system. An experimental application of the human lifting motion was carried out to demonstrate the effectiveness of the proposed approach. The results illustrated that the proposed algorithm can achieve increased simplicity, improved effectiveness, good robustness, and a higher precision of computation.