Context-based selection and execution of robot perception graphs

  title={Context-based selection and execution of robot perception graphs},
  author={Nico Hochgeschwender and Miguel Angel Olivares-M{\'e}ndez and Holger Voos and Gerhard K. Kraetzschmar},
  journal={2015 IEEE 20th Conference on Emerging Technologies \& Factory Automation (ETFA)},
To perform a wide range of tasks service robots need to robustly extract knowledge about the world from the data perceived through the robot's sensors even in the presence of varying context-conditions. This makes the design and development of robot perception architectures a challenging exercise. In this paper we propose a robot perception architecture which enables to select and execute at runtime different perception graphs based on monitored context changes. To achieve this the architecture… 

Figures and Tables from this paper

Configuration of Perception Systems via Planning Over Factor Graphs

This paper model the configuration of perception systems as a planning problem over probabilistic graphical models, specifically the used sensors, data processing algorithms and view poses, and works on a subset of the full configuration space of perceptions systems.

Graph-based software knowledge: Storage and semantic querying of domain models for run-time adaptation

This work argues that the robots themselves should be first-class consumers of this knowledge to autonomously adapt their software to the various and changing run-time requirements induced, for instance, by the robot's tasks or environment.

Automatic Configuration of the Structure and Parameterization of Perception Pipelines

This paper introduces a structure template based on hypothesis generation, hypothesis refinement, and hypothesis testing to demonstrate the effectiveness of the approach to perform automatic configuration based on structure templates and sequential model-based optimization.

A concept for declarative information acquisition in smart environments

A new and embedded programming language that combines both an SQL-like syntax with new semantics, and declarative aspects with imperative scripting capabilities has been developed that integrates these technologies from different layers, which allows to combine and access them freely.

Selecting Adequate Aerial Perceptual Functions with Fuzzy Logic

A method to predict the performance of perception functions, allowing a ranking for algorithm selection, using expert knowledge to model the influence of the environment on the perception functions using fuzzy logic is introduced.

Performance prediction and selection of aerial perception functions during UAV missions

An algorithm selection method using fuzzy inference systems (FIS) trained on example data to predict the perception functions performance under different environmental conditions is proposed to maintain the overall performance for surveillance missions.

VISUAL 2016 Proceedings

An ontology-based system to inventory and model installed sensor and respective data processing resources on-board airborne surveillance aircrafts and it is possible to identify and manage processing chains capable of solving complex image processing tasks.

Towards Persistent Storage and Retrieval of Domain Models using Graph Database Technology

This work uses graph database technology to persistently store and retrieve robot domain models and develops a graph database system that can be used to manage and retrieve these models.

Marker based attendance systems in education process

This paper describes a simply use case, to do students attendance using only classroom live video stream and a simply organized list of optical markers, and used Aruco markers as marker detection method / system.



Declarative Specification of Robot Perception Architectures

The paper presents the domain-specific language RPSL, some examples of its application, the current state of implementation and some validation experiments, and suggests to specify the integral parts of robot perception architectures using explicit models, which allows to easily configure, modify, and validate them.

INVICON: A Toolkit for Knowledge-Based Control of Vision Systems

The case studies show that INVICON reduces the effort needed to build KB controllers for challenging domains and improves their flexibility and robustness.

Girgit: A Dynamically Adaptive Vision System for Scene Understanding

Girgit is presented, a context-aware vision system for scene understanding, that dynamically orchestrates a set of components that encapsulates a vision-related algorithm such as from the OpenCV library.

Automatic parameter regulation of perceptual systems

  • D. Hall
  • Computer Science
    Image Vis. Comput.
  • 2006

Conceptual Spaces — The Geometry of Thought

Conceptual Spaces—The Geometry of Thought is a book by Peter Gärdenfors, professor of cognitive science at Lund University, Sweden, a definitive account of the widely examined AGM theory of belief revision.

Autonomic Computer Vision Systems

This paper uses a software component model to describe techniques for regulation of internal parameters, error detection and recovery, self-configuration and self-repair for vision systems.

Run Time Adaptation of Video-Surveillance Systems: A Software Modeling Approach

This paper describes a framework for design, deployment, and run-time adaptation of video-surveillance systems--with a focus on the run time aspect, and uses Feature Diagrams which offer a convenient way of representing the variability of a software system.

Domain Specific Languages

A number of DSLs spanning various phases of software development life cycle in terms of features that elucidates their advantages over general purpose languages and perform in depth study by practically applying a few open source DSLs: ‘Cascading’, Naked Objects Framework and RSpec.

Domain Specific Languages, 1st ed

  • Addison-Wesley Professional,
  • 2010