Humanoid Mobile Manipulation Using Controller Refinement

Abstract

An important class of mobile manipulation problems are "move-to-grasp" problems where a mobile robot must navigate to and pick up an object. One of the distinguishing features of this class of tasks is its coarse-to-fine structure. Near the beginning of the task, the robot can only sense the target object coarsely or indirectly and make gross motion toward the object. However, after the robot has located and approached the object, the robot must finely control its grasping contacts using precise visual and haptic feedback. This paper proposes that move-to-grasp problems are naturally solved by a sequence of controllers that iteratively refines what ultimately becomes the final solution. This paper introduces the notion of a refining sequence of controllers and defines it in terms of controller goal regions and domains of attraction. Refining sequences are shown to be more robust than other types of controller sequences. In addition, a procedure for converting a refining sequence into an equivalent "parallelized" controller is proposed. Executing this parallelized controller confers all the advantages of iteratively executing the controllers sequentially. The approach is demonstrated in a move-to-grasp task where Robonaut, the NASA/JSC dexterous humanoid, is mounted on a mobile base and navigates to and picks up a geological sample box.

EL-E: an assistive mobile manipulator that autonomously fetches objects from flat surfaces

“…The bung hole is only visible when the robot is close to the drum, so it first navigates with respect to the drum, and then, once the bung hole is visible, navigates up to it and inserts the probe. Similarly, Platt et al (2006) use a three step turn-drive-turn control policy for a robot approaching and grasping a box. They show that this multi-step policy is beneficial, since the variance in the robot's estimate of the position of the box decreases as the robot gets closer.…”

Dual arm manipulation—A survey

“…Most of these methods have been traditionally aimed at a single manipulator. While it is possible, in some cases, to trivially extend these methods to two arms, just by repeating the process for each arm, there are some peculiarities of dual-arm manipulation which require some modifications to the frameworks [101] [168].…”

Learning to Manipulate Articulated Objects in Unstructured Environments Using a Grounded Relational Representation

“…Whereas approaches in the previous category are concerned with individual controllers, approaches in the third category sequence [3,18] or compose [20] multiple controllers to generate more complex manipulation behaviors. Composite controllers can be arranged into state transition diagrams to further increase robustness and versatility [19].…”