Situation assessment for human-robot interactive object manipulation

Abstract

In daily human interactions spatial reasoning occupies an important place. With this ability we can build relations between objects and people, and we can predict the capabilities and the knowledge of the people around us. An interactive robot is also expected to have these abilities in order to establish an efficient and natural interaction. In this paper we present a situation assessment reasoner, based on spatial reasoning and perspective taking, which generates on-line relations between objects and agents in the environment. Being fully integrated to a complete architecture, this reasoner sends the generated symbolic knowledge to a fact data base which is built on the basis on an ontology and which is accessible to the entire system. This work is also part of a broader effort to develop a complete decisional framework for human-robot interactive task achievement.

Combining symbolic and geometric planning to synthesize human-aware plans: toward more efficient combined search

“…Their representation changes depending on the planner: at the symbolic level, attributes are associated to an entity in the form of predicates, whereas at the geometric level the same entity is described by its shape and its configuration space. The geometric reasoning system is able to compute spatial relations between entities, such as between objects (in, ontop-of, ...) or between agents and objects (visibility, reachability, ...) [18]. This can be related to well known need to deal with the anchoring problem in order to fill the gap between the levels [19].…”

A Reactive Controller Based on Online Trajectory Generation for Object Manipulation

“…More information about human-aware motion planning in the system can be found in [23]. More about geometrical reasoning can be found in [27], [24], and [28]. Physical Human Robot Interaction is a dynamic research area including various aspects, including safety, control architecture, planning, human intention recognition, and more.…”

“…The 3D cost maps used are of different types: collision risk map calculated based on the minimum distance between trajectory and the obstacles; visibility and reachability map of a human [27] and safety and comfort 3D map of a human, Figure 3 shows two examples of cost maps. For example, when the risk of collision with the robot base is high, the trajectory can be controlled in robot the frame.…”

Grounding the Interaction: Knowledge Management for Interactive Robots

“…In this section we present SPARK (SPAtial Reasoning & Knowledge [121]), a situation assessment reasoner that generates symbolic knowledge from the geometry of the environment with respect to relations between objects, robots and humans, also taking into account the different perspective that each agent has on the environment. SPARK can be seen as an amodal geometric model of the environment that serves both as basis for the fusion of the perception modalities and as bridge with the symbolic layer.…”