Work

Planning and estimation algorithms for human-like grasping

Year: 2016
Type: dissertation
Abstract: The use of robots in human-like environments requires them to be able to sense and model unstructured scenarios. Thus, their success will depend on their versatility for interacting with the surroundings. This interaction often includes manipulation of objects for accomplishing common daily tasks. Therefore, robots need to sense, understand, plan and perform; and this has to be a continuous loop.\n\t\t\t\t This thesis presents a framework which covers most of the phases encountered in a common manipulation pipeline. First, it is shown how to use the Fast Marching Squared algorithm and a leader-followers strategy to control a formation of robots, simplifying a high dimensional path-planning problem. This approach is evaluated with simulations in complex environments in which the formation control technique is applied. Results are evaluated in terms of distance to obstacles (safety) and the needed deformation.\n\t\t\t\t Then, a framework to perform the grasping action is presented. The necessary techniques for environment modelling and grasp synthesis and path planning and control are presented. For the motion planning part, the formation concept from the previous chapter is recycled. This technique is applied to the planning and control of the movement of a complex hand-arm system. Tests using robot Manfred show the possibilities of the framework when performing in real scenarios.\n\t\t\t\t Finally, under the assumption that the grasping actions may not always result as it was previously planned, a Bayesian-based state-estimation process is introduced to estimate the final in-hand object pose after a grasping action is done, based on the measurements of proprioceptive and tactile sensors. This approach is evaluated in real experiments with Reex Takktile hand. Results show good performance in general terms, while suggest the need of a vision system for a more precise outcome.
Author: David Álvarez-Sánchez
Language: English
Open Access status: green
OpenAlex ID: W2613879085
Cites: 149
Cited by: 1
Related to: 10