Automated kinematic synthesis of planar mechanisms with revolute joints

Kinematic synthesis of mechanisms at both the type and dimensional levels is a major challenge in the field of mechanism design. This paper presents a new approach for tackling this issue using the framework of genetic algorithms (GAs). The genome employed for a pin-connected planar mechanism includes two parts: a resizable joint-link incidence matrix representing the topology, and three variable-length real-number arrays storing geometric parameters, initial conditions, and task specific parameters. A multibody analysis engine based on an absolute coordinate formulation returns the kinematic simulation results for each genome. The GA can explore different types of mechanisms by applying genetic operators to selected genomes. The evolutionary process is not dependent on the results obtained from enumeration of mechanisms. Case studies on mechanism path generation problems show the applicability of this approach to the kinematic synthesis of pin-connected planar mechanisms.