Things are made for what they are: Solving manipulation tasks by using functional object classes

Abstract

Solving arbitrary manipulation tasks is a key feature for humanoid service robots. However, especially when tasks involve handling complex mechanisms or using tools, a generic action description is hard to define. Different objects require different handling methods. Therefore, we try to solve manipulation tasks from point of view of the object, rather than in the context of the robot. Action templates within the object context are introduced to resolve object specific task constraints. As part of a centralized world representation, the action templates are integrated into the planning process. This results in an intuitive way of solving manipulation tasks. The underlying architecture as well as the mechanisms are discussed within this paper. The proposed methods are evaluated in two experiments.

Collaborative Autonomy between High‐level Behaviors and Human Operators for Remote Manipulation Tasks using Different Humanoid Robots

“…In this approach, a tablet application for shared autonomy between a semiautonomous robot and a remote operator is presented. This approach is based on previous work designed to provide a fully autonomous robot with symbolic and geometric information of the objects required to perform a manipulation task (Leidner, Borst, & Hirzinger, ). Although this approach provides an interesting concept on how to ground symbolic information of a task into robot actions, it does not provide an interface to define these motions with respect to a point of interest in the object grasped.…”

A knowledge-driven shared autonomy human-robot interface for tablet computers

“…In our previous work we have shown that it is vital to arrange knowledge within an object-centered context to solve everyday manipulation tasks [22]. Following this concept, objects are organized in a hierarchical structure according to their functionality.…”

“…Geometric backtracking in case of unsuccessful simulation is inherent in this step. Please refer to [22] for a more detailed description. This approach has been successfully evaluated to solve everyday manipulation tasks [22], mobile manipulation tasks [24], and force-sensitive whole-body manipulation [25].…”

Command Robots from Orbit with Supervised Autonomy

“…Task level command has shown its viability in teleoperation tasks which require high levels of dexterity [4]. Supervised autonomy differ from telepresence in that the commanded robot utilizes a blend of local intelligence and prior object knowledge to reason the appropriate action to complete each commanded task [3] . This also enables coping with time-delays of multiplesecond in the communication link, which is not possible for full haptic feedback telepresence.…”