| Though robots have been used in manufacturing industries for decades in precise and repetitive tasks, nowadays they are entering the realm of household environments. While the ”dream” is of an autonomous robotic butler, still the present scope is limited to only vacuum cleaners and lawn mowers, due to the limited capability specifically in the field of grasping and manipulation in the unstructured environment. The thesis approaches towards a goal of grasping novel objects using low-cost and noisy sensors with an incomplete object model, by incorporating a superellipsoid modeling. A complete grasping system is proposed relying on real-time superellipsoid representation of partial-view objects and incomplete ob-ject modeling, well suited for unknown symmetric objects in cluttered scenario followed by optimized antipodal grasping. The incomplete object models are processed through a mir-roring algorithm that assumes symmetry to first create an approximate complete model and then fit for superquadric representation. The grasping algorithm is designed for maximum force balance and stability, taking advantage of the quick retrieval of dimension and surface curvature information from the superquadric parameters. The pose of the superquadric with respect to the direction of gravity is calculated and used together with the parameters of the superquadric and specification of the gripper, to select the best direction of approach and contact points. The grasp hypotheses are filtered by incorporating a reachability map ensuring the execution of grasps. The methods are evaluated on custom datasets contain-ing objects in isolation as well as in clutter on three different robotic platforms a) a 7DOF WAM arm, b) a 6DOF UR5 arm with 2-finger gripper and c) a PR2 mobile manipulator. Though the method is based on simplistic shape and geometry information, it outperforms existing state-of-art learning-based algorithms in cluttered environments measured against time efficiency and accuracy.Key Words: Grasp Planning, Unknown Objects, Superellipsoid, Symmetry model |