Planning in-hand object manipulation with multifingered hands considering task constraints

Abstract

In-hand manipulation with a multifinger hand is defined as changing the object pose from an initial to a final grasp configuration, while maintaining the fingertip contacts on the object surface. Given only the task constraints, represented as a desired motion of the object and an external force to be applied or resisted, the problem can be expressed as finding a good set of contact points on the object and a corresponding hand configuration compatible with the task to be executed. This paper presents a method for solving such problem, taking into account the kinematic structure and torque limits of the hand, the force closure condition (which must be guaranteed during the whole trajectory), and task compatibility. The feasibility of such method is tested in simulation of 2D and 3D examples.

In-hand manipulation using gravity and controlled slip

“…Han et al proposed an in-hand manipulation framework that combines rolling and finger gaiting [5]. Hertkorn et al formulated a planning framework which also takes into consideration kinematic and dynamic constraints of the task [6]. Okamura et al formulated a survey of different dexterous manipulation techniques that have been proposed in the literature, as well as a summary of the main kinematic, contact and dynamic models used in those techniques [7].…”

“…(6). Even though the mass and center of mass can be estimated online by the robot just before running the controller by using a force-torque sensor, this estimate is subject to measurement errors arising from e.g.…”

Learning robot in-hand manipulation with tactile features

“…Because generally we are working with samples from reference distribution q, the weighting factors w should be normalized by q: w(a, s) = π(a|s)µ π (s)/q(a, s). The parameter η, that specifies the transformation and determines how greedy the optimization is, is a Lagrangian parameter that is naturally set through optimization of the dual function (6). The resulting weighting factors are…”

“…Such approaches have been used successfully for finger gaiting [2], manipulation with a rolling contact [2][3][4][5], sliding [5], in-grasp-manipulation [6], and rolling objects over the hand palm [7]. Optimization techniques have also been used for a variety of in-hand motions [8].…”

“…These models can be used to apply methods from planning, control and optimization to manipulate known objects [2][3][4][5][6][7][8]. To manipulate unknown objects as well, systems could be designed to reactively adapt to the object rather than relying on knowing the object's model.…”