engleski

Optimization of Chebyshev’s Lambda Mechanism stride length by combining Genetic algorithm and numerical simulation

Walking mechanisms are linkages that imitate the walking motion of a human or animal leg, and are considered when wheeled drives are not applicable, mostly on uneven or stepped surfaces. Depending on the number of linkages and joints, as well as the desired leg trajectory, several types of mechanisms were developed. For each type, the leg trajectory is affected by linkage lengths ; therefore, default lengths are provided to ensure stable walking motion. In this paper, Chebyshev’s Lambda mechanism was optimized to increase the step length, which was selected as the sole objective function. Linkage lengths were varied, while Grashof’s law was used to ensure the functionality of each proposed solution. The sum of linkage lengths was kept constant to allow for a meaningful comparison between the proposed solutions. The genetic algorithm (GA) was used to generate, vary, and update populations. To obtain the objective function values, Simscape Multibody (a MATLAB Simulink subroutine) analysis was embedded within the GA to numerically determine the positions of significant points within the mechanism for a full cycle. As a result, a feasible mechanism with the largest stride length was obtained.

Chebyshev’s Lambda Mechanism, walking mechanism, optimization, genetic algorithm

nije evidentirano