Improving grasp quality for 3D objects using Particle Swarm Optimization (PSO) and Mesh Parameterization

Handling of objects through robot hands is a complex problem in most applications. This makes grasp planning very important and thereby it becomes necessary to formulate efficient schemes for automated grasp synthesis. A method for fast synthesis of grasp configurations is proposed in this work through the use of Particle Swarm Optimization (PSO). The algorithm uses a feasible grasp as the input to improve the grasp quality. The largest ball criterion is used as a measure of the grasp quality. Surface tessellated objects are used as test cases for the implementation and the optimization has been conducted for frictional and non-frictional instances.