Developing genetic programming techniques for the design of compliant mechanisms

Abstract.

Compliant mechanisms achieve desired force and displacement characteristics through elastic deformation of their structure. Current research in the synthesis of compliant mechanism topology has pursued multiobjective optimization using gradient-based search methods. This paper will explore the use of a random-guided search method for multiobjective optimization of compliant mechanisms through genetic programming techniques. The combination of genetic algorithms and compliant mechanisms is an effective and interesting combination of two biologically inspired engineering design areas. This paper will describe and demonstrate the successful use of genetic programming to create a general design tool for topological synthesis of compliant mechanisms. Features that exploit the implementation of genetic algorithms to compliant mechanism design, such as multiple criteria specification, multiple-design parameter variation, and final selections from a family of solutions will be presented and discussed. Finally, the use of this design tool will be demonstrated on several familiar examples for validation and discussion.