• Corpus ID: 17582553

Affordance-Based Reasoning in Robot Task Planning

@inproceedings{Awaad2013AffordanceBasedRI,
  title={Affordance-Based Reasoning in Robot Task Planning},
  author={Iman Awaad and Gerhard K. Kraetzschmar},
  year={2013}
}
Humans are able to come up with plans to achieve their goals, and to adapt these plans to changes in their environment, finding fixes, alternatives and taking advantages of opportunities without much deliberation. For example, they may use a tea kettle to water the plants, or a mug instead of a glass. Despite decades of research, artificial agents are not as robust or as flexible. In this work, we introduce three reasoning phases that use affordances to enable such robustness and flexibility in… 

Figures from this paper

Planning Domain + Execution Semantics: A Way Towards Robust Execution?
TLDR
It is shown that the combined used of causal, temporal and categorical knowledge allows the robot to detect failures even when the effects of actions are not directly observable.
A Formal Model of Affordances for Flexible Robotic Task Execution
TLDR
A novel conceptualization of affordances and its realization as a description logic ontology that proposes candidate affordances through inference and that these can then be validated through physics-based simulation.
Generating and Executing Hierarchical Mobile Manipulation Plans
TLDR
The integration of an off-the-shelf HTN planner into a robotic system which is aimed at enabling a service robot to learn from experiences and a state-machine-based approach is employed.
Affordance-feasible planning with manipulator wrench spaces
TLDR
This work investigates mechanical wrenches as a viable representation of these affordance requirements and capabilities and proposes decomposing a behavior into a vocabulary of characteristic requirements and capability that are suitable to predict the affordances of various parts of the workspace.
Towards Human-Level Semantics Understanding of Human-Centered Object Manipulation Tasks for HRI: Reasoning About Effect, Ability, Effort and Perspective Taking
TLDR
It is argued that such symbolic level understanding of a task, which is not bound to trajectory, kinematics structure or shape of the robot, facilitates generalization to novel situations as well as ease the transfer of acquired knowledge among heterogeneous robots.
Affordance Research in Developmental Robotics: A Survey
TLDR
This survey provides an entry point for interested researchers, including a general overview of affordance research, classification and critical analysis of existing work, and discussion of how affordances are useful in developmental robotics.
Leveraging Large-Scale Semantic Networks for Adaptive Robot Task Learning and Execution
TLDR
This article shows that the task context provided by the task labels within the action structure of a task plan can be leveraged to disambiguate information within a noisy large-scale semantic network containing hundreds of potential object candidates to identify successful object substitutions with high accuracy.
Adapting Everyday Manipulation Skills to Varied Scenarios
TLDR
The solution encodes tool-using manipulation skills in a generic way, with parameters that can be filled in at run-time via queries to a robot perception module, which abstracts the functional parts of the tool and extracts key parameters that are needed for the task.
A deep probabilistic framework for heterogeneous self-supervised learning of affordances
TLDR
A flexible deep probabilistic framework is proposed which allows an explorative agent to learn tool-object affordances in continuous space by incorporating various probability distributions tailored for each individual modality and by eliminating the need for any pre-processing of the data.
Similarity Reasoning over Semantic Context-Graphs
TLDR
This dissertation leverages general purpose semantic networks to derive similarity measures in a problem-independent manner and demonstrates that the task context is essential in establishing which objects are interchangeable.
...
1
2
...

References

SHOWING 1-10 OF 16 REFERENCES
Fusing DL Reasoning with HTN Planning
  • R. Hartanto
  • Computer Science
    KI - Künstliche Intelligenz
  • 2010
TLDR
A novel approach is introduced which amalgamates Description Logic (dl) reasoning with Hierarchical Task Network (htn) planning, where the planning domain description as well as fundamental htn planning concepts are represented in dl and can therefore be subject to dl reasoning.
Integration of Online Learning into HTN Planning for Robotic Tasks
TLDR
The idea behind this work is to use the HTN domain to constrain the space of possibilities, and then to learn on the constrained space in a way requiring few training samples, rendering the method applicable to autonomous mobile robots.
Affordance-Based Planning for Assisting Humans in Daily Activities
TLDR
A functional representation of everyday objects, places and actions described in terms of affordances is proposed, to represent affordances and capabilities as regions in a proper affordance and capability space, and to describe such regions using neural maps.
A General Approach to Synthesize Problem-Specific Planners
TLDR
A way to improve the performance of hand-tailorable planners by compiling each domain description into a separate domain-specific planner is described.
Flexible Strategy Learning: Analogical Replay of Problem Solving Episodes
TLDR
The effectiveness of the analogical replay strategy is demonstrated by providing empirical results on the performance of a fully implemented system, PRODIGY/ANALOGY, accumulating and reusing a large case library in a complex problem solving domain.
Probabilistic policy reuse in a reinforcement learning agent
TLDR
Interestingly and almost as a side effect, Policy Reuse also identifies classes of similar policies revealing a basis of core policies of the domain, demonstrating that such a basis can be built incrementally, contributing the learning of the structure of a domain.
Representing actions and functional properties in conceptual spaces
The book Conceptual Spaces (Gardenfors 2000) presents a theory of concepts based on geometrical and topological structures in spaces that are built up from “quality dimensions”. Most of the examples
Technology affordances
TLDR
This discussion illustrates this discussion with several examples of interface techniques, and suggests that the concept of affordances can provide a useful tool for user-centered analyses of technologies.
Cognitive, physical, sensory, and functional affordances in interaction design
  • H. Hartson
  • Computer Science
    Behav. Inf. Technol.
  • 2003
TLDR
This article argues that the concept of physical affordance carries a mandatory component of utility or purposeful action (functional affordance) and provides guidelines to help designers think about how these four kinds of affordance work together naturally in contextualized HCI design or evaluation.
Conceptual Space Markup Language (CSML): Towards the Cognitive Semantic Web
  • B. Adams, M. Raubal
  • Computer Science
    2009 IEEE International Conference on Semantic Computing
  • 2009
TLDR
The design decisions behind CSML are described, the key component elements of a CSML document are introduced, and examples of its usage are presented.
...
1
2
...