Robotic manipulation in unstructured environments requires grasping a wide range of objects. Tactile sensing is presumed to provide essential information in this context, but there has been little work examining the tactile sensor signals produced during realistic manipulation tasks. This paper presents tactile sensor data from grasping a generic object in thousands of trials. Position error between the hand and object was varied to model the uncertainty in real-world grasping, and a grasp outcome prediction was done using only tactile sensors. Results show that tactile signals are highly variable despite good repeatability in grasping conditions. The observed variability appears to be intrinsic to the grasping process, due to the mechanical coupling between fingers as they contact the object in parallel, as well as numerous factors such as frictional effects and inaccuracies in the robot hand. Using a simple machine learning algorithm, grasp outcome prediction based purely on tactile sensors is not reliable enough for real-world responsibilities. These results have implications for improved tactile sensor system and controller design, as well as signal processing and machine learning methods.

Grasping and manipulation in unstructured environments must handle a wide range of object properties and significant sensing errors. Underactuation and compliance have been shown to be an effective way to improve grasping performance under such uncertainty, but the degree of compliance plays an important role in both gently adapting to sensing errors and maintaining stable grasps of heavy objects. These demands limit the range of objects that can be grasped. We consider the role and required characteristics of tactile sensing as a compensation method when compliance alone is insufficient. By strategic use of contact sensing, it is possible to expand the capabilities of a hand to grasp effectively under a wide range of positioning errors using simple position-driven motors and low-cost hardware.

Micromanipulation methods used for complicated tasks such as microrobot assembly and microvascular surgery often lack the force reflection and contact localization capability necessary to achieve robust grasps of micro-scale objects without applying excessive forces. This absence of haptic feedback is especially prohibitive in cases where visual evidence of force application, such as object surface deformation, is imperceptible and where unstructured, dynamically changing environments require force sensing and modulation for safe, atraumatic object manipulation. This paper describes the design, fabrication, and experimental validation of a soft tactile sensor array for sub-millimeter contact localization and contact force measurement during micromanipulation. The geometry and placement of conductive liquid embedded channels within the sensor array are optimized to provide adequate sensitivity for representative micro-manipulation tasks. Mechanical testing of the sensor demonstrates a sensitivity of less than 50mN and contact localization resolution on the order of 100's of microns.