We developed a soft robotic gripper that is capable of robustly greedy and identifying objects based totally on interior country measurements along with a gadget that autonomously performs grasps. The incredibly compliant smooth hand allows for intrinsic robustness to greedy uncertainties. The addition of inner sensing allows the configuration of the hand and object to be detected. The finger module includes resistive pressure sensors on the fingertips for contact detection and resistive bend sensors for measuring the curvature profile of the finger. The curvature sensors can be used to estimate the contact geometry. This capability allows to distinguish between a set of grasped objects. With one information factor from each finger, the object can be recognized by using greedy it. A clustering algorithm finds the correspondence for every grasped object. We tested this for each enveloping grasps and pinch grasps. A closed loop machine uses a digital camera to realize approximate object locations. Compliance in the tender hand handles that uncertainty in addition to geometric uncertainty in the form of the object.