Defect-free optimal synthesis of crank-rocker linkage using nature-inspired optimization algorithms

This paper proposes the synthesis of a human knee exoskeleton using the reduced number of necessary and sufficient constraints. The proposed human knee exoskeleton consists of a defect-free crank-rocker mechanism capable of accurately moving its coupler point along the prescribed trajectory. For synthesizing the crank-rocker mechanism based on proposed reduced number of constraints, an optimization problem is formulated to synthesize the mechanism. An algorithm based on teaching-learning-based-optimization (TLBO) is presented to solve this highly nonlinear optimization problem. The optimization minimizes the error between generated and prescribed trajectory and simultaneously avoids any defect in the synthesized mechanism. The penalty method is used to manage all the constraints. Besides a realistic nontrivial example of human knee flexion/extension, a straight line trajectory is also considered to demonstrate the effectiveness of the refinement scheme in the optimal syntheses of planar crank-rocker linkage free from all defects. The optimization problem is solved using well-established nature-inspired algorithms such as genetic algorithm (GA), particle swarm algorithm (PSO), and teaching-learning-based-optimization (TLBO) algorithm. It is found that TLBO is computationally more efficient than other algorithms used here. Additionally, the proposed human knee exoskeleton model is experimentally validated for one gait cycle.